DPMV 3 Operations Guide
DPMV 3 Operations Guide
Feedback
Send suggestions and comments about this document to [email protected].
This document is provided “as-is”. Information and views expressed in this document, including
URL and other Internet Web site references, may change without notice. You bear the risk of
using it.
Some examples depicted herein are provided for illustration only and are fictitious. No real
association or connection is intended or should be inferred.
This document does not provide you with any legal rights to any intellectual property in any
Microsoft product. You may copy and use this document for your internal, reference purposes.
You may modify this document for your internal, reference purposes.
© 2010 Microsoft Corporation. All rights reserved.
Microsoft, Active Directory, Hyper-V, SQL Server, Windows, Windows PowerShell, Windows
Server, and Windows Vista are trademarks of the Microsoft group of companies.
All other trademarks are property of their respective owners.
Contents
DPM 2010 Operations..................................................................................................................21
In This Section...........................................................................................................................21
Locating Information.....................................................................................................................41
See Also....................................................................................................................................42
Using Migrate-Datasource............................................................................................................61
Syntax.......................................................................................................................................62
Things to Remember.................................................................................................................63
Using MigrateDatasourceDataFromDPM......................................................................................63
Syntax.......................................................................................................................................64
Examples...............................................................................................................................65
Dismounting Databases................................................................................................................76
Recovering Mailboxes...................................................................................................................88
In This Section...........................................................................................................................89
Protecting an Exchange Server 2007 SCR Target Server Configured as Single Node Cluster....96
Using ConfigureSharePoint........................................................................................................123
Permissions.............................................................................................................................123
Syntax.....................................................................................................................................124
Parameters...........................................................................................................................124
Upgrading SharePoint.................................................................................................................155
SQL Server Hardware Upgrade...............................................................................................155
When Instances of SQL Server Do Not Change......................................................................156
Protecting Hyper-V.....................................................................................................................173
DPM Support Scenarios..........................................................................................................173
Unsupported Scenarios...........................................................................................................173
See Also..................................................................................................................................174
Conditions When DPM Fails to Back up Hyper-V Virtual Machines in an Online State..............177
Considerations for Backing Up Virtual Machines on CSV with Hardware VSS Providers...........199
Considerations for Backing Up Virtual Machines on CSV with the System VSS Provider..........201
Enabling Per Node Serialization..............................................................................................201
Enabling Per CSV LUN Serialization.......................................................................................201
Procedure to Create the DataSourceGroups.xml File and Serialize the Backup Jobs............203
Procedure to Generate the DataSourceGroups.xml File on a CSV Cluster.............................203
Procedure to Merge the DataSourceGroups.xml Files from All CSV Clusters.........................204
Using DPMSync..........................................................................................................................222
Managing Performance...............................................................................................................224
In This Section.........................................................................................................................224
See Also..................................................................................................................................224
Replica Creation.........................................................................................................................225
See Also..................................................................................................................................227
Change Tracking.........................................................................................................................227
See Also..................................................................................................................................227
Synchronization..........................................................................................................................227
See Also..................................................................................................................................228
Consistency Check.....................................................................................................................228
See Also..................................................................................................................................228
Backup to Tape...........................................................................................................................229
See Also..................................................................................................................................229
DPM Processes..........................................................................................................................229
See Also..................................................................................................................................229
Performance Counters................................................................................................................230
See Also..................................................................................................................................232
Improving Performance...............................................................................................................232
In This Section.........................................................................................................................232
See Also..................................................................................................................................232
Modifying Workloads...................................................................................................................232
In This Section.........................................................................................................................233
See Also..................................................................................................................................233
Increasing Capacity....................................................................................................................237
See Also..................................................................................................................................237
Managing Tapes..........................................................................................................................241
In This Section.........................................................................................................................241
Short Erase.................................................................................................................................241
Enabling Short Erase...............................................................................................................241
In This Section
Managing DPM Servers
Managing Protected File Servers and Workstations
Managing Protected Servers Running Exchange
Managing Protected Servers Running SQL Server
Managing Protected Servers Running SharePoint
Managing Protected Virtual Servers
Managing Protected Client Computers
Managing Hyper-V Computers
Managing Protected Computers in Workgroups and Untrusted Domains
Managing System Protection
Managing Performance
Managing Tapes
Appendix A: Quick Reference to DPM Tasks
Appendix B: DPM 2010 Schema Extension
Appendix C: Custom Report Views
Appendix D: Configuration of Inbox Tracing
Appendix E: Windows Server Logo Certification
21
Managing DPM Servers
As a system administrator, you are accustomed to managing servers in different roles. You plan
your maintenance routines to accommodate each server’s role, and you take that role into
account when making structural changes such as changing the server name or relocating the
server. So what do you need to consider when the role of a server running System Center Data
Protection Manager (DPM) is added to your network structure?
This section discusses performing common maintenance tasks on DPM servers. It provides
guidance on making changes to server configurations after DPM is set up and on how DPM
manages time zones. This section provides information about configuring firewalls on both the
DPM server and protected computers so that communication can be maintained. This section
also provides recommendations for monitoring DPM and offers methods for monitoring.
In This Section
Performing General DPM Server Maintenance
Performing DPM Server Management Tasks
Managing the Storage Pool
Monitoring DPM Server
See Also
Managing Performance
Managing Protected File Servers and Workstations
Managing Protected Servers Running Exchange
Managing Protected Servers Running SQL Server
Managing Protected Servers Running SharePoint
Managing Protected Virtual Servers
In This Section
Using Windows Maintenance Tools on the DPM Server
Applying Operating System Updates to the DPM Server
22
Running Antivirus Software on the DPM Server
See Also
Managing the Storage Pool
Monitoring DPM Server
Performing DPM Server Management Tasks
23
copies may be lost when you defragment a
volume (http://go.microsoft.com/fwlink/?
LinkId=65210).
Chkdsk.exe: Use to check the file system and Do not run chkdisk on DPM replica and
file system metadata for errors and to display a recovery point volumes. Chkdsk causes the
status report of its findings. volumes to dismount, and if data is written to
the replica volume while the recovery point
volume is dismounted, it might cause a
complete loss of recovery points.
See Also
Applying Operating System Updates to the DPM Server
Running Antivirus Software on the DPM Server
Using Windows Maintenance Tools on File Servers and Workstations
Using Windows Maintenance Tools on SQL Servers
Using Windows Maintenance Tools on Exchange Servers
Using Windows Maintenance Tools on SharePoint Servers
Using Windows Maintenance Tools on Virtual Server
24
You can subscribe to Microsoft Update at any time on the Microsoft Update Web site
(http://go.microsoft.com/fwlink/?LinkId=41291).
See Also
Running Antivirus Software on the DPM Server
Using Windows Maintenance Tools on the DPM Server
25
cleaning and quarantining can result in data corruption because these processes cause the
antivirus software to modify files, making changes that DPM cannot detect.
Whenever DPM attempts to synchronize a replica that has been modified by another
program, data corruption of the replica and recovery points can result. Configuring the
antivirus software to delete infected files resolves this problem. For information about
configuring your antivirus software to delete infected files, see the documentation for your
antivirus software.
I
m
p
o
r
t
a
n
t
You must run a manual synchronization with consistency check job each time that the
antivirus software deletes a file from the replica, even though the replica will not be
marked as inconsistent.
See Also
Applying Operating System Updates to the DPM Server
Using Windows Maintenance Tools on the DPM Server
In This Section
Managing the DPM Database Volume
Finding DPM Servers in Active Directory Domain Services
How to Migrate a DPM Server to New Hardware
Restarting the DPM Server
26
Moving the DPM Server to a New Domain
Renaming the DPM Server
Changing the SQL Server Instance Used by DPM
Coordinating Protection Across Time Zones
How to Change the Time Zone of the DPM Server
See Also
Managing the Storage Pool
Monitoring DPM Server
Performing General DPM Server Maintenance
See Also
Performing DPM Server Management Tasks
27
No
te
If DPM is installed on a server that is not a member of a domain and the server is then
added to a domain, the service connection point will not be registered in Active Directory
Domain Services.
To locate DPM servers in Active Directory Domain Services, use a query tool such as Adsiedit to
find all computers in the domain that have a “serviceClassName=MSDPM” service connection
point.
No
te
Adsiedit is a Microsoft Management Console (MMC) snap-in that is available when you
install the Windows Server 2003 Support Tools. For more information about using
Adsiedit, see Adsiedit Overview on the Windows Server 2003 TechCenter
(http://go.microsoft.com/fwlink/?LinkId=50377).
To
ins
tall
Wi
nd
ow
s
Se
rve
r2
00
3
su
pp
ort
too
ls
28
To
loc
ate
DP
M
ser
ver
s
by
usi
ng
Ad
sie
dit
1. Run adsiedit.msc.
2. Right-click the Domain node, point to New, and then click Query.
3. Enter a name for the query, such as “MSDPM Servers.”
4. Choose the Machines node as the root of the search.
5. In Query String, enter serviceClassName=MSDPM.
6. Click OK to display a query node under the Domain node.
7. Select the query node; the servers on which DPM is installed are displayed in the list
pane.
See Also
Performing DPM Server Management Tasks
29
To
mi
gra
te
a
DP
M
ser
ver
to
ne
w
har
dw
are
,
1. Install DPM on a new server. For more information, see Installing DPM 2010
(http://go.microsoft.com/fwlink/?LinkID=179156).
2. Identify a protected computer to migrate and run SetDPMServer.exe -DPMServerName <Name
of new DPM server> on the protected computer.
N
o
t
e
See Also
Performing DPM Server Management Tasks
30
Restarting the DPM Server
If you need to restart the DPM server for any reason, check the Monitoring task area in DPM
Administrator Console for jobs currently running, and then follow these guidelines:
• If there are no jobs currently running or scheduled to run during the time required for the
restart, restart the DPM server.
• If a synchronization with consistency check job is running, restart the DPM server.
Synchronization with consistency check will resume at the next scheduled time or you can
retry the job manually.
• If a replica creation job is running, postpone the restart until the job is completed. If the restart
cannot be postponed, you must run synchronization with consistency check manually for the
replica after you restart the DPM server.
• If any synchronizations or express full backups are scheduled to run during the restart, either
postpone the restart until the recovery points are created or re-run the synchronizations and
create the recovery points manually after you restart the DPM server.
• If any jobs that use the tape library are running, postpone the restart until the jobs are
complete. If the restart cannot be postponed, the following job types will be canceled by the
restart and must be re-run after the restart:
• Back up to tape
• Copy to tape
• Recovery from tape
• Tape verification
• If you are erasing a tape, postpone the restart until the current job is complete. Cancel any
pending tape erase jobs, restart the computer, and then reschedule the canceled tape erase
jobs.
See Also
Performing DPM Server Management Tasks
See Also
Performing DPM Server Management Tasks
31
Renaming the DPM Server
You cannot rename a DPM server.
See Also
Performing DPM Server Management Tasks
See Also
Performing DPM Server Management Tasks
32
in the time zone of the DPM server. Although you schedule jobs to run in the time zone of the
protected computer, the start times and recovery point times of the jobs are displayed in the time
zone of the DPM server.
For example, suppose that your DPM server is located in Berlin and a protected file server is
located in Reykjavik, which is two hours earlier than Berlin. When you schedule synchronization
and the recovery point for 6:00 P.M., the jobs run at 6:00 P.M. in Reykjavik time, the time on the
file server. However, if a user in Reykjavik requests to have data recovered to its state as of
6:00 P.M. yesterday, you must search for the recovery point that represents 8:00 P.M. Berlin time,
because the DPM recovery user interface represents recovery point times in the time zone of the
DPM server.
In DPM Administrator Console, in the Recovery task area, the Last Modified column displays the
date and time of the most recent changes to the file, which could be either changes to the
contents or changes to the metadata.
Work hours for network bandwidth usage throttling use the time zone of the protected computer.
33
protected server resides in a location that does not—the start of daylight saving time disrupts the
time zone offsets between DPM and the protected computer.
To resolve this problem, you can force the DPM server to reset the time zone offset by removing
the data sources from protection and then adding the data sources back to protection groups.
See Also
How to Change the Time Zone of a File Server or Workstation
How to Change the Time Zone of the DPM Server
Performing DPM Server Management Tasks
To
ch
an
ge
the
tim
e
zo
ne
of
the
DP
M
ser
ver
34
See Also
How to Change the Time Zone of a File Server or Workstation
Coordinating Protection Across Time Zones
Performing DPM Server Management Tasks
N
o
t
e
The subnet should cover the entire range of network addresses for the DPM server
and the servers you intend to protect.
3. Restart the DPM agent on the DPM server and the protected computers. It may cause
ongoing tasks to fail. Post a restart, watch out for alerts, and perform the recommended
actions, if needed.
Example
This example details the process of setting up a backup network address for a DPM server
protecting another server. All names and addresses are hypothetical and for illustration only.
35
The existing backup setup consists of dpm.x.y.com protecting ps.x.y.com. Name lookup using
“nslookup” on either server returns the following IPs (that is, each IP address is visible to each
node):
No
te
The name lookups must be performed on the FQDNs; for example, “nslookup
ps.x.y.com”.
Now, to set up a backup network, another NIC is added to each of the above servers and
connected to another network such as 192.168.1.0/24 with a corresponding subnet mask
255.255.255.0. When the network and NICs are configured, the name lookup using “nslookup”
returns two addresses per server as given below.
We recommend that you verify whether the DPM server is able to ping the protected computer’s
backup network address (192.168.1.24). Similarly, the protected computer should be able to ping
the DPM server’s backup network address (192.168.1.23).
At this stage, backup LAN configuration information is added to the DPM server as follows:
Add-BackupNetworkAddress -DpmServername DPM -Address 192.168.1.0/24
-SequenceNumber 1
No
te
36
No
te
No
te
The FQDN and version of DPM on both Server1 and Server2 must be the same.
37
To back up the DPM database
1. At the command prompt, run DPMBackup.exe -db, located at Microsoft Data Protection
Manager\DPM\bin.
2. In the console tree of the backup program, browse to Microsoft Data Protection
Manager\DPM\Volumes\ShadowCopy\Database Backups. The file name of the DPM
database backup is DPMDB.bak.
3. Select the media to which you want to back up the database.
4. Start the backup.
To install DPM
For information about how to install DPM, see Installing DPM 2010
(http://go.microsoft.com/fwlink/?LinkId=179156).
1. At the command prompt, type DpmSync –restoredb –dbloc <DPMDB location>, and
then press ENTER.
N
o
t
e
38
4. After you restart the failed jobs, you must perform a consistency check for all data
sources. For more information about how to perform a manual consistency check, see
"How to synchronize a replica" in DPM 2010 Help.
Remove-ProductionServer.PS1
Syntax: Remove-ProductionServer.ps1 -DPMServername [DPMServerName] -PSName
[ProtectedComputerName]
Parameter Description
-DPMServername Name of the DPM server.
-PSName Name of the protected computer that must be
removed.
If the computer was protected using FQDN or
NETBIOS name, you must use that name to
here.
Im
po
rta
nt
There should be no actively protected data sources on the computer you are trying to
remove.
39
Managing the Storage Pool
The storage pool is a set of disks on which the DPM server stores the replicas and recovery
points for the protected data. DPM can use any of the following for the storage pool:
• Direct attached storage (DAS)
• Fibre Channel storage area network (SAN)
• iSCSI storage device or SAN
The storage pool supports most disk types, including Integrated Drive Electronics (IDE), Serial
Advanced Technology Attachment (SATA), and SCSI, and it supports both the master boot record
(MBR) and GUID partition table (GPT) partition styles.
You cannot add USB/1394 disks to the DPM storage pool.
DPM cannot use space in any pre-existing volumes on disks added to the storage pool. Although
a pre-existing volume on a storage pool disk might have free space, DPM can use space only in
volumes that it creates. To make the entire disk space available to the storage pool, delete any
existing volumes on the disk and then add the disk to the storage pool.
Im
po
rta
nt
In This Section
Adding Disks to the Storage Pool
How to Replace a Disk in the Storage Pool
Removing a Disk from the Storage Pool
See Also
Monitoring DPM Server
Performing DPM Server Management Tasks
Performing General DPM Server Maintenance
40
Adding Disks to the Storage Pool
DPM cannot use space in any pre-existing volumes on disks added to the storage pool. Although
a pre-existing volume on a storage pool disk might have free space, DPM can use space only in
volumes that it creates. To make the entire disk space available to the storage pool, delete any
existing volumes on the disk and then add the disk to the storage pool.
DPM regularly rescans the disks and volumes in the storage pool and updates the storage pool
space. If you add a disk that contains a volume to the storage pool and later delete that volume,
when DPM rescans the disk, it will add the new unallocated space to the available storage pool.
If the name of a disk is listed as “Unknown” on the Disks tab in the Management task area of
DPM Administrator Console, you cannot add the disk to the storage pool until the disk name is
corrected. To resolve this issue, perform the following procedure.
To
cor
rec
ta
dis
k
na
me
N
o
t
e
All disks without a friendly name are listed as “Disk Drive.” An example of a
friendly name is “HITACHI_DK23EB-40”.
3. On the Action menu, click Scan for hardware changes to reinstall the disk.
See Also
How to Replace a Disk in the Storage Pool
Removing a Disk from the Storage Pool
41
How to Replace a Disk in the Storage Pool
You can use the following procedure to replace a disk in the storage pool if a disk fails.
To
rep
lac
ea
dis
k
in
the
sto
rag
e
po
ol
1. In the Disk Management console, identify the replica volumes and recovery point
volumes that are stored on the failed disk.
2. Remove protection from the data sources that have replica volumes and recovery point
volumes on the failed disk, and select Delete protected data.
3. Physically remove the disk that needs to be replaced.
4. Physically add the replacement disk.
5. In DPM Administrator Console, click Management on the navigation bar, and then click
the Disks tab.
6. Select the disk that you removed, and in the Actions pane, click Remove.
7. In the Actions pane, click Add.
8. In the Available disks section, select the replacement disk, click Add, and then click OK.
9. Add the data sources from step 2 to an existing protection group, or create a new
protection group for these data sources.
a. If you create a new protection group and have tape backup of the data sources,
create the replicas manually by using the tape backup.
b. If you create a new protection group and do not have tape backup of the data
sources, allow DPM to create the replicas across the network.
c. If you add the data sources to an existing protection group, DPM will start an
immediate consistency check, which will re-create the replicas.
42
N
o
t
e
See Also
Adding Disks to the Storage Pool
Removing a Disk from the Storage Pool
See Also
Adding Disks to the Storage Pool
How to Replace a Disk in the Storage Pool
43
Monitoring DPM Server
After you set up data protection, you should monitor DPM activity to verify that everything is
working correctly and to troubleshoot any problems that occur. Monitoring is essential to give you
an overview of what has already happened, what is currently happening, and what is scheduled
to happen. By monitoring DPM, you will know that data protection activities are working as
expected, and you will have confidence that errors and warnings will be brought to your attention
when they occur.
No
te
In This Section
Establishing a Monitoring Schedule
Locating Information
Methods for Monitoring DPM
See Also
Managing Performance
Managing the Storage Pool
Performing DPM Server Management Tasks
Performing General DPM Server Maintenance
44
Daily • Critical and warning alerts Replica issues, synchronization
• E-mail notifications (if they and recovery point creation
are configured) issues, agent issues, jobs
waiting for tape, backup failures
• Status report
Monthly Reports: Trends and patterns that might
• Status indicate problems or potential
• Tape Management issues
• Disk Utilization
On Demand Recovery job status Recovery job failures
See Also
Locating Information
Managing Performance
Methods for Monitoring DPM
Locating Information
After you implement your monitoring schedule, you will observe certain trends and notice various
alerts. You might want to investigate the issues underlying the alerts, troubleshoot problems, or
analyze some of the trends. DPM provides a number of resources to help you with your research.
The following table lists a number of references that you can use to locate information that will
help you answer many common questions.
Information Locations
What do you want to know? Look here:
Does anything need my attention? • E-mail notifications of alerts, if you
Are there any changes on the protected subscribe to them
computers that affect data protection? • Monitoring task area, Alerts tab
Did all the backups that were supposed to • Status report
happen yesterday happen correctly? • Protection report
Is there an issue that keeps coming up?
Are recovery goals being met?
Do I need to add disk space to the storage • Management task area, Disks tab
pool? • Disk Utilization report
When will a job run? Monitoring task area, Jobs tab
How long did the last consistency check take?
How much data was transferred by the most
recent synchronization job?
45
How many recovery points are available for a • Protection task area, Details pane
data source? • Recovery task area
Are all replicas consistent?
What tapes are available in the library? Management task area, Libraries tab
What data is on each tape?
Did a recovery job complete successfully? • Monitoring task area, Alerts tab
• Monitoring task area, Jobs tab
• E-mail notification (if you subscribe to e-
mail notification when you initiate a
recovery)
Is the DPM server able to contact each Management task area, Agents tab
protected computer?
What is the status of the DPM service? • Microsoft Management Console (MMC)
Services snap-in
• Event log, in case of service failures
What problems have occurred over the past • Status report
month? • Monitoring task area, Alerts tab, with
Show inactive alerts selected
What is the status of each of my DPM servers MOM Operator console, State view
and the computers that they protect?
Why is recovery point creation failing for a Status report
protection group member?
See Also
Establishing a Monitoring Schedule
Managing Performance
Methods for Monitoring DPM
46
• Monitor operations for multiple DPM servers by using the System Center Data Protection
Manager (DPM) Management Pack for Microsoft Operations Manager 2005 or System
Center Operations Manager 2007.
• Monitor the instance of SQL Server that DPM installs by using the System Center
SQL Server Management Pack for Microsoft Operations Manager 2005.
In This Section
Monitoring with DPM Administrator Console
Monitoring with Reports and Alert Notifications
Monitoring with DPM Management Packs
See Also
Establishing a Monitoring Schedule
Locating Information
Managing Performance
No
te
You do not need to monitor each task area in DPM Administrator Console. For more
information, see Establishing a Monitoring Schedule.
47
For monitoring purposes, the Alerts tab provides the more critical information. You should check
the Alerts tab daily to provide timely resolution of issues that might be preventing successful
protection of data.
48
N
o
t
e
49
Management Task Area
The Management task area contains three tabs: Disks, Agents, and Libraries.
50
You can enable the DPM reporting feature at any time after installing and configuring DPM.
However, to ensure that DPM has enough information to generate meaningful report data, we
recommend that you wait at least a day after starting data protection activities to begin viewing
reports. For instructions to help you enable DPM reporting, see Using Reports
(http://go.microsoft.com/fwlink/?LinkId=196780) in DPM 2010 Help.
No
te
When a DPM server is protecting a large number of computers, you should stagger the
delivery schedule for reports sent by e-mail. If you schedule all reports to be sent at the
same time, the memory limitations of SQL Server Reporting Services might prevent some
reports from being sent.
The following table summarizes the available reports and indicates how you should use them. For
information about interpreting the data in reports, see Report Types
(http://go.microsoft.com/fwlink/?LinkId=196781) in DPM 2010 Help.
DPM Reports
Report Name Summary of Contents
51
Status The Status report provides the status of all recovery
points for a specified time period, lists recovery jobs,
and shows the total number of successes and failures
for recovery points and disk-based and tape-based
recovery point creations. This report shows trends in the
frequency of errors that occur and lists the number of
alerts.
Use this report to answer questions such as the
following:
• What happened yesterday? Last week? Last
month?
• What succeeded and what failed?
• What is the trend of errors? Which errors occur most
frequently?
• Are we achieving the recovery point objective (RPO)
established in our service level agreement (SLA)?
No
te
52
metrics for recovery success rolled up over long periods
of time to track how recoveries are doing.
Use this report to identify how well you performed
against your service level agreements for recovery time
objectives and recovery success guarantees.
Disk Utilization Summarizes disk capacity, disk allocation, and disk
usage in the DPM storage pool.
Use this report to do the following:
• Identify trends in disk usage
• Make decisions about modifying space allocations
for protection groups and adding disks to the
storage pool
• Identifying how much disk resource each computer
is using on DPM
See Also
Managing Performance
Monitoring DPM Server
Monitoring with DPM Management Packs
Monitoring with Reports and Alert Notifications
53
these subject lines when you set up rules in Outlook to filter reports and alert notifications into
specific folders. You can customize your e-mail notifications by using Operations Manager.
See Also
Managing Performance
Monitoring DPM Server
Monitoring with DPM Administrator Console
Monitoring with DPM Management Packs
54
From the Operations Manager server, administrators can perform the following monitoring tasks
for managed DPM servers and the computers that they protect:
• Centrally monitor the health and status of data protection and critical performance indicators
of multiple DPM servers and the computers that they protect.
• View the state of all roles on DPM servers and computers servers.
• Monitor actionable DPM alerts relating to replica creation, synchronization, and recovery point
creation. The DPM Management Pack filters out alerts that do not require an action, such as
a synchronization job in progress.
• Through Operations Manager alerts, monitor the status of memory, CPU, and disk resources
on DPM servers, and be alerted to DPM database failures.
• Monitor resource usage and performance trends on DPM servers.
• Diagnose and resolve problems on a remote DPM server.
The DPM Management Packs are not included with the DPM product. You can download
management packs at the DPM Management Pack download site
(http://go.microsoft.com/fwlink/?linkid=50208).
See Also
Managing Performance
Monitoring DPM Server
Monitoring with DPM Administrator Console
Monitoring with Reports and Alert Notifications
55
attributes of the protected volume. For example, in the following scenario, if you recover File1, or
File1 and Folder1, to an alternate destination folder named Test, the Test folder acquires the
attributes of Volume X.
VolumeRoot (X:)
\Folder1
\Folder2
File2
File1
After performing the recovery, set the appropriate permissions for Folder1.
56
DPM Setup fails if a pre-release version of .NET
Framework 2.0 is already installed
If a pre-release version of .NET Framework 2.0 is already installed on the computer on which you
are installing DPM 2010, Setup fails with an "unhandled exception" when installing .NET
Framework 2.0.
Before running DPM 2010 Setup, uninstall the pre-release version of .NET Framework 2.0 by
using Add or Remove Programs from Control Panel. Then reinstall DPM, and DPM Setup will
install the current version of .NET Framework 2.0.
57
DPM always defaults to port 25 when sending
alerts through the SMTP server
In DPM 2010, if you specify any port number in the SMTP server settings other than 25, DPM will
continue to default to port 25 to communicate with the SMTP server.
58
Managing Protected File Servers and
Workstations
The topics in this section provide information about performing common maintenance tasks on
protected file servers and workstations, as well as guidance for making changes to the computer
configuration or cluster configuration after the computer or cluster is protected by System Center
Data Protection Manager (DPM) 2010.
In This Section
Performing General Maintenance on File Servers and Workstations
Performing File Server and Workstation Management Tasks
Managing Clustered File Servers
Additional Resources
Disaster Recovery (http://go.microsoft.com/fwlink/?LinkID=179152)
See Also
Managing DPM Servers
Managing Performance
Managing Protected Servers Running Exchange
Managing Protected Servers Running SQL Server
Managing Protected Servers Running SharePoint
Managing Protected Virtual Servers
Managing Tapes
59
No
te
If you disable a protection agent for a server that is a cluster node, you should disable the
protection agent for every node of the cluster.
To
dis
abl
ea
pr
ote
cti
on
ag
ent
In This Section
Using Windows Maintenance Tools on File Servers and Workstations
Applying Operating System Updates on File Servers and Workstations
Running Antivirus Software on File Servers and Workstations
See Also
Managing Clustered File Servers
Performing File Server and Workstation Management Tasks
60
Using Windows Maintenance Tools on File
Servers and Workstations
In general, you can continue maintenance on file servers and workstations protected by DPM
using your regular maintenance schedule and the maintenance tools provided in the operating
system. Those tools and any impact on data protection are listed in the following table.
See Also
Applying Operating System Updates on File Servers and Workstations
Managing Protected File Servers and Workstations
Running Antivirus Software on File Servers and Workstations
Using Windows Maintenance Tools on the DPM Server
Using Windows Maintenance Tools on Exchange Servers
Using Windows Maintenance Tools on SQL Servers
Using Windows Maintenance Tools on SharePoint Servers
61
Using Windows Maintenance Tools on Virtual Server
See Also
Managing Protected File Servers and Workstations
Running Antivirus Software on File Servers and Workstations
Using Windows Maintenance Tools on File Servers and Workstations
See Also
Applying Operating System Updates on File Servers and Workstations
Managing Protected File Servers and Workstations
Using Windows Maintenance Tools on File Servers and Workstations
62
Performing File Server and Workstation
Management Tasks
When events or business requirements demand it, you might need to make changes to your
protected file servers and workstations or to the data sources on the protected computer. The
topics in this section discuss the impact certain changes might have on DPM protection.
In This Section
Changing the Path of a Data Source
Moving File Servers and Workstations Between Domains
How to Rename a File Server or Workstation
How to Change the Time Zone of a File Server or Workstation
Using Migrate-Datasource
Using MigrateDatasourceDataFromDPM
See Also
Managing Clustered File Servers
Managing Performance
Performing General Maintenance on File Servers and Workstations
63
See Also
How to Change the Time Zone of a File Server or Workstation
How to Rename a File Server or Workstation
Managing Protected File Servers and Workstations
Moving File Servers and Workstations Between Domains
To
ch
an
ge
the
do
ma
in
me
mb
ers
hip
of
a
pr
ote
cte
d
co
mp
ute
r
64
If you retain the replicas and recovery points, the data will remain accessible for
administrative recovery until you delete the replicas. However, it will not be accessible for
end-user recovery.
2. Uninstall the protection agent by using DPM Administrator Console on the DPM server.
3. Change the domain membership of the computer.
4. Install a protection agent by using DPM Administrator Console on the DPM server.
5. Add the data sources to protection groups on the DPM server.
For information about performing tasks involving protection agents and protection groups,
see DPM Help.
See Also
Changing the Path of a Data Source
How to Change the Time Zone of a File Server or Workstation
How to Rename a File Server or Workstation
Managing Protected File Servers and Workstations
65
To
ren
am
ea
pr
ote
cte
d
co
mp
ute
r
1. Stop protection for all data sources on the computer by removing them from the
protection group.
If you retain the replicas and recovery points, the data will remain accessible for
administrative recovery until you delete the replicas. However, it will not be accessible for
end-user recovery.
2. Uninstall the protection agent by using DPM Administrator Console on the DPM server.
3. Change the name of the computer.
4. Install a protection agent by using DPM Administrator Console on the DPM server.
5. Add the data sources to protection groups on the DPM server.
For information about tasks that involve protection agents and protection groups, see
DPM Help.
See Also
Changing the Path of a Data Source
How to Change the Time Zone of a File Server or Workstation
Managing Protected File Servers and Workstations
Moving File Servers and Workstations Between Domains
66
For more information about time zones and DPM protection, see Coordinating Protection Across
Time Zones.
To
up
dat
e
the
tim
e
zo
ne
in
the
DP
M
dat
ab
as
e
See Also
Changing the Path of a Data Source
How to Rename a File Server or Workstation
Managing Protected File Servers and Workstations
Moving File Servers and Workstations Between Domains
67
Using Migrate-Datasource
Migrate-Datasource is a command-line script that lets you continue protecting a data source (file,
folder, volume, or share) to the same replica volume even after it has been migrated to a different
volume on the same protected computer. You have to run the Migrate-Datasource script even if
you have not changed the drive letters of the volume because DPM recognizes volumes by the
GUID and not the drive letter.
Im
po
rta
nt
If you have secondary DPM protection configured, you must run the Migrate-Datasource
script on the secondary server also.
No
te
Syntax
Migrate-Datasource.ps1 [-DPMServerName] <string> [-Option [auto or manual]] [-PSName]
<string>
Parameter Description
DPMServerName Name of the DPM server from which you want
to migrate data.
Option Indicates whether DPM should perform an
automatic or manual migration.
Automatic: If you specify the automatic option,
DPM migrates all the data sources on the
68
protected computer. Use this option if you
created a new volume but retained the logical
path (for example, if F:\ was reformatted on a
new disk but is still called F:\, or the mount
point is still the same). DPM automatically
updates the mappings for the replica of F:\ to
the new volume that is now called F:\.
Manual: If you specify the manual option, you
have to migrate each data source individually.
The script gives you a list of volumes protected
by DPM that are not present and a list of
unprotected volumes. You can then map the
volumes individually. Use this option if the
logical path was not preserved (for example,
G:\ became H:\).
PSName Name of the protected computer to which the
data source is being migrated.
Things to Remember
• Migrate-Datasource is used only for migration of file system data sources, such as volumes.
For other data sources, follow the instructions in the alerts.
• DPM does not support migration from a volume on a drive (for example, D:\) to a mounted
volume (for example, E:\<mountpoint>, where mountpoint is a location on which the volume
has been mounted).
• For auto-migration of mounted volumes, the volume on the new computer should have the
same mount point name as the volume on the previously protected computer. DPM does not
allow you to migrate to a drive.
• For migration of mounted volumes (where the old volume is protected by using a mount
point):
• If the protected volume has multiple mount points, at least one mount point of the volume
on the new computer should have the same mount point path as before.
Old volume: C:\mnt
new volume: C:\mnt (may have drive letter and other mount points)
• If the volume also has a drive letter, only the drive letter is visible while you select the
new volume for migration. This should be selected manually.
• You should migrate volumes only if you have reformatted them or if the volume GUID
associated with the volume has changed.
69
No
te
After migration, you cannot perform original location recovery for the recovery points
created before the migration. Recovery fails with the message Couldn't find the
selected volume. You can, however, recover to an alternate location.
Using MigrateDatasourceDataFromDPM
MigrateDatasourceDataFromDPM is a command-line script that lets you migrate DPM data for a
data source – replica volumes and recovery point volumes – across disks. Such a migration might
be necessary when your disk is full and cannot expand, your disk is due for replacement, or disk
errors show up.
No
te
70
Syntax
MigrateDatasourceDataFromDPM.ps1 [-DPMServerName] <string> [-Source] <disk[]> [-
Destination] <disk[]>
MigrateDatasourceDataFromDPM.ps1 [-DPMServerName] <string> [-Source] <data source> [-
Destination] <disk[]>
MigrateDatasourceDataFromDPM.ps1 [-DPMServerName] <string> [-Source] <data source> [-
Destination] <DPM server volume[]>
Parameter Description
DPMServerName Name of the DPM server for which you want to
migrate data.
Source The location from which the data must be
moved. This can be either a DPM disk (use
Get-DPMDisk to retrieve the disk) or a DPM
data source (use Get-Datasource to retrieve
the data source). The source can be a set of
disks.
Destination The location to which the data must be moved.
This can be either a DPM disk array (use Get-
DPMDisk to retrieve the array of disks) or an
array of two DPM volumes (use Get-
DPMVolume to retrieve the list of DPM volumes
on the server). The first element of the array is
the destination replica volume and the second
the recovery point volume.
The destination disks must be added to the
DPM disk pool before migration.
No
te
Examples
The following examples show how the script works.
Example 1: Disk D1 contains the replica and recovery points for the data source DS1.
You need to retain D1 for the retention range of the latest recovery point on it, usually one month.
After the latest recovery point expires, DPM will deallocate the replicas and recovery points on
disk D1 automatically.
Example 2: Disk D1 contains the replica of the data source DS1 and the recovery point for data
source DS2. Disk 2 contains the replica of DS2 and the recovery point for DS1.
If you do a DPM disk to DPM disk migration to a third disk (Disk 3), this disk will have four
volumes, replicas and recovery points for DS1 and DS2.
$disk = Get-DPMDisk –DPMServerName DPMTestServer
./MigrateDatasourceDataFromDPM.ps1 –DPMServerName DPMTestServer –Source
$disk[0] –Destination $disk[2]
This results in the following:
This happens because DPM cannot move just a replica or a recovery point, it will always move
them in pairs, hence even though the command only moves the volumes from D1, DPM will move
also the related replica and recovery point.
Example 3: Disk D1 contains the replica of the data source DS1 and the recovery point for data
source DS2. Disk D2 contains the replica of DS2 and the recovery point for DS1.
72
If you choose to migrate only the data for DS1 to a third disk (Disk 3), this disk will have two
volumes, the replica and recovery point for DS1.
$pg = Get-ProtectionGroup DPMTestServer
$ds = Get-Datasource $pg[0]
$disk = Get-DPMDisk –DPMServerName DPMTestServer
./MigrateDatasourceDataFromDPM.ps1 –DPMServerName DPMTestServer –Source $ds[0]
–Destination $disk[2]
In This Section
Changing File Server Cluster Members
Changing Resource Groups on Clustered File Servers
See Also
Managing Performance
Performing File Server and Workstation Management Tasks
Performing General Maintenance on File Servers and Workstations
73
For example, assume you have a server cluster that contains four computers: Node1, Node2,
Node3, and Node4. You need to replace computer Node4 with a new computer, named Node5.
You use the administration console for your cluster service to add Node5 to the cluster and
configure the resources that can be failed over to Node5.
DPM issues an alert that protection of the server cluster will fail until a protection agent is installed
on Node5. You install the protection agent on Node5.
You fail over the resources from Node4 to other nodes in the cluster. When no resources remain
on Node4, you remove it from the cluster. DPM detects the failovers and continues protection of
the cluster.
DPM detects that Node4 has left the cluster—it appears as a stand-alone node now. If it no
longer exists on the network, you can remove the record for this server in DPM Administrator
Console.
See Also
Changing Resource Groups on Clustered File Servers
Performing File Server and Workstation Management Tasks
Performing General Maintenance on File Servers and Workstations
74
See Also
Changing File Server Cluster Members
Performing File Server and Workstation Management Tasks
Performing General Maintenance on File Servers and Workstations
In This Section
What's New for Protecting Exchange Server 2010
Performing General Maintenance on Servers Running Exchange
Performing Exchange Server Management Tasks
Managing Clustered Exchange Servers
Recovering Exchange Data
Managing Exchange SCR Servers
In This Section
Exchange Server 2010 Prerequisites
Installing Protection Agents on Exchange Server 2010 Nodes
Protecting Exchange Server 2010
75
Recovering Exchange Server 2010 Data
No
te
When you install a protection agent on a DAG node, DPM 2010 displays the following
warning: "You cannot protect cluster data in the selected nodes without installing agents
on the other nodes." This is a DPM 2010 warning when you are protecting clusters. This
does not relate to Exchange Server 2010 and you can ignore this message.
76
No
te
When parallel backups are tried on multiple copies of same Exchange Server database,
then backups will fail.
In addition to the wizard pages you used to protect Exchange Server 2007, you perform the steps
on the following wizard pages to protect Exchange Server 2010:
1. On the Select Protection Group Type page, select Server, and then click Next to continue.
2. On the Select Group Members page, expand the domain under which the DAG resides,
expand the DAG, all the existing databases together with their respective nodes are
displayed. Select the data you want to protect, and then click Next to continue.
N
o
t
e
The Create New Protection Group Wizard does not indicate which databases are
active or passive. You must already know which databases are active or passive
when performing the steps in the wizard. For servers that are part of a DAG, the
databases will be listed under the <DAG-name> node.
Insert art: Databases in a DAG that have been selected for protection.
3. On the Specify Exchange Protection Options page, specify if you want to run the Eseutil
tool on one of the Exchange Server databases. For members on Exchange Server 2010,
select if you want to run the Eseutil tool for both the database and the log files or just for the
log files. If you are protecting DAG servers, Exchange Server recommends that you run the
Eseutil tool for log files only. For stand-alone servers, we recommend that you select both the
database and the log files.
4. On the Specify Exchange DAG Protection page, select the databases for copy backup and
express full backup. For protecting multiples copies of the same database, select only one
database for express full and incremental backup and then select the remaining copies for
copy backup.
Insert art: Databases selected for express full and copy backups.
5. On the Select Data Protection Method page, select whether you want to use short-term
disk-based protection or long-term tape-based protection, and then click Next to continue.
6. On the Select Short-term Goals page, specify your protection goals such as retention range
and synchronization frequency, and then click Next to continue.
77
7. On the Summary page, review your selections, and then click Create Group to complete the
wizard.
N
o
t
e
After creating the protection group for the Exchange Server database, changing the
status of the Exchange Server database from active to passive or vice-versa on the
Exchange Server does not require any changes on the DPM server. DPM will
continue backing up the data from the same node without any failures.
78
1. Recover the database to its original location. Overwrite the existing copy of the database.
2. Recover the database to an alternate database. Restore to another database on an
Exchange Server.
3. Recover to an Exchange Recovery database. Recover to an Exchange Recovery database
instead of a standard mailbox database.
4. Recover to network location. Copy the database to a network folder.
5. Copy to tape. Create an on-tape copy of the database.
No
te
DPM 2010 does not support recovering mailbox databases to passive databases. While
recovering to the original database or to an alternate database, the target database on
which the recovery is being performed should not be passive.
No
te
If you disable a protection agent for a server that is a cluster node, you should disable the
protection agent for every node of the cluster.
79
To
dis
abl
ea
pr
ote
cti
on
ag
ent
In This Section
Using Windows Maintenance Tools on Exchange Servers
Performing Exchange Maintenance Tasks
Applying Operating System Updates on Exchange Servers
Running Antivirus Software on Exchange Servers
See Also
Using Windows Maintenance Tools on the DPM Server
Using Windows Maintenance Tools on File Servers and Workstations
Using Windows Maintenance Tools on SQL Servers
Using Windows Maintenance Tools on SharePoint Servers
Using Windows Maintenance Tools on Virtual Server
80
Performing Exchange Maintenance Tasks
Most Microsoft Exchange maintenance tasks should have no adverse affect on performance or
data protection. However, special considerations apply when you are performing offline database
defragmentation on Exchange servers that are protected by Data Protection Manager (DPM).
Offline defragmentation involves using the Exchange Server Database Utilities (Eseutil.exe), an
Exchange Server utility that you can use to defragment, repair, and check the integrity of
Exchange server databases.
If you must perform an offline defragmentation, you should perform a synchronization with
consistency check for protected storage groups when defragmentation is complete.
81
Performing Exchange Server Management
Tasks
This section provides instructions and guidelines for managing a protected Exchange server and
making changes after the initial DPM configuration.
In This Section
Upgrading Exchange Server 2003 to Exchange Server 2007
Moving Exchange Servers Between Domains
How to Rename an Exchange Server
Adding Storage Groups and Databases
Dismounting Databases
Changing the Path of a Database or Log File
Renaming Storage Groups
Moving Databases Between Storage Groups
82
How
to
ma
int
ain
dat
a
pr
ote
cti
on
du
rin
ga
tra
nsi
tio
n
to
Ex
ch
an
ge
Se
rve
r2
00
7
83
Moving Exchange Servers Between Domains
You cannot do the following for protected computers:
• Change the domain of a protected computer and continue protection without disruption.
• Change the domain of a protected computer and associate the existing replicas and recovery
points with the computer when it is re-protected.
We recommend that you do not change the domain of a protected computer. If you must change
the domain of a protected computer, you must complete two tasks:
• Remove the data sources on the computer from protection while the computer retains its
original domain membership.
• Protect the data source on the computer after it becomes a member of another domain.
84
To resolve this issue, you must manually reconfigure an express full backup on another copy of
the database so that log truncation happens for that database.
Dismounting Databases
When a database that belongs to a protected storage group is dismounted, protection jobs for
that database only will fail. Logs for that storage group will not be truncated. However, the longer
that the database remains dismounted, the more likely it is that the log space on the Microsoft
Exchange server will overflow, which will result in the dismount of the storage group on the
Exchange server. If the database will not be needed, you should delete it.
85
• Recover the storage group to a network folder as files.
To
ch
an
ge
the
na
me
of
a
pr
ote
cte
d
sto
rag
e
gr
ou
p
86
Moving Databases Between Storage Groups
The following table describes the impact on data protection when you move a database between
storage groups.
87
tbl_RM_RecoverySource
Rebuilding Indexes
Rebuilding an index deletes the index and creates a new one. Rebuilding an index removes
fragmentation and reclaims disk space by compacting the pages that are using the specified or
existing fill factor setting, and the index rows are reordered in contiguous pages, allocating new
pages as needed. This can improve SQL query performance by reducing the number of page
reads required to obtain the requested data.
Query to rebuild an index
USE DPMDB
GO
ALTER INDEX ALL ON <tableName> REBUILD
GO
Reorganizing Indexes
Reorganizing an index defragments the leaf level of clustered and nonclustered indexes on tables
and views by physically reordering the leaf-level pages to match the logical order (left to right) of
the leaf nodes. Having the pages in order improves index-scanning performance. The index is
reorganized within the existing pages allocated to it; no new pages are allocated. If an index
spans more than one file, the files are reorganized one at a time. Pages do not migrate between
files.
Reorganizing an index also compacts the index pages. Any empty pages created by this
compaction are removed providing additional available disk space. In some cases, the gain might
not be significant. It is also takes longer than rebuilding the index.
Query to rebuild indexes
USE DPMDB
GO
ALTER INDEX ALL ON <tableName> REORGANIZE
GO
88
Not a time intensive operation. Usually a time intensive operation.
Most effective when index is heavily Most effective when index is not heavily
fragmented. fragmented.
Additional Resources
ALTER INDEX (Transact-SQL) (http://go.microsoft.com/fwlink/?LinkId=129339)
In This Section
Changing Exchange Server Cluster Members
Changing Resource Groups on Clustered Exchange Servers
89
Changing Resource Groups on Clustered
Exchange Servers
A cluster node can have any number of resource groups. Moving a protected data source to a
resource group, between resource groups, or out of a resource group can cause protection job
failures. To successfully make any of those changes to resource group membership, perform the
following steps:
1. Stop existing protection of the data source. The data source could belong to a protection
group as a single data source on a protected server or as a data source as a member of a
resource group.
2. Begin protection of the data source according to its new status, either as a single data source
on a protected server or as a data source as a member of a resource group. This will allocate
a new replica for the data source.
Changing the name of a resource group will affect the protection of all data sources in the
resource group. To change the name of a resource group, perform the following steps:
1. Stop protection of the resource group.
2. Change the name of the resource group.
3. Begin protection of the resource group under its new name.
90
This option is not available if you select Latest as the recovery point.
• Copy the database to a network folder.
This option is not available if you select Latest as the recovery point. Data Protection
Manager (DPM) creates the following directory structure at the destination that you specify:
DPM_Recovery_Point_timestamp\DPM_Recovered_At_timestamp\Server name\Exchange
application\Database name\Files
To use the Bring the database to a clean shutdown after copying the files option, the
DPM protection agent and the Eseutil utility must be installed on the destination server. The
Eseutil utility can be installed as part of either an Exchange Server installation or an
Exchange Server Administrator-only-mode installation.
• Copy the database to tape.
This option is not available if you select Latest as the recovery point. This option copies the
replica of the storage group that contains the selected database.
In This Section
How to Recover a Storage Group to its Original Location
How to Recover a Database to Its Original Location
How to Recover a Database to an Alternate Database
How to Copy Exchange Data to a Network Folder
How to Copy Exchange Data to Tape
Recovering Mailboxes
Recovering Data to Clustered Servers
91
How
to
rec
ov
er
a
sto
rag
e
gr
ou
p
to
its
ori
gin
al
loc
ati
on
1. On the server to which the storage group will be recovered, configure each database to
allow it to be overwritten by the recovered data. For instructions, see "How to Configure
the Exchange Databases so That the Restore Process Overwrites Them"
(http://go.microsoft.com/fwlink/?LinkId=97929).
2. In DPM Administrator Console, click Recovery on the navigation bar.
3. Using the browse functionality, select the storage group to recover.
4. On the calendar, click any date in bold to obtain the recovery points available for that
date. The Recovery time menu lists the time for each available recovery point.
5. On the Recovery time menu, select the recovery point you want to use.
6. In the Actions pane, click Recover.
The Recovery Wizard starts. The wizard options vary depending on the version of
Exchange.
7. On the Review recovery selection page, click Next.
8. Select Recover to original Exchange Server location, and then click Next.
9. On the Specify recovery options page, you can select Send an e-mail when this
recovery completes.
Select this option to specify an e-mail address or addresses to notify upon recovery
completion. If you select this option, you must enter the e-mail address to notify. Multiple
e-mail addresses must be separated by a comma.
10. On the Summary page, review the recovery settings, and then click Recover.
See Also
How to Recover a Database to Its Original Location
92
How to Recover a Database to an Alternate Database
How to Copy Exchange Data to a Network Folder
How to Copy Exchange Data to Tape
Recovering Mailboxes
Recovering Data to Clustered Servers
No
te
In Exchange 2007, if there are multiple databases in a storage group, all databases will
be dismounted during recovery. An Exchange 2007 best practice is to have one database
per storage group.
93
How
to
rec
ov
er
a
dat
ab
as
e
to
its
ori
gin
al
loc
ati
on
1. On the server to which the database will be recovered, configure the target database to
allow it to be overwritten by the recovered data. For instructions, see "How to Configure
the Exchange Databases so That the Restore Process Overwrites Them"
(http://go.microsoft.com/fwlink/?LinkId=97929).
2. In DPM Administrator Console, click Recovery on the navigation bar.
3. Using the browse functionality, select the database to recover.
4. On the Recovery time menu, select Latest.
You must select the most recent recovery point to recover the storage group to its original
location.
5. In the Actions pane, click Recover.
The Recovery Wizard starts. The wizard options will vary depending on the version of
Exchange.
6. On the Review recovery selection page, click Next.
7. Select Recover to original Exchange Server location, and then click Next.
8. On the Specify recovery options page, you can select Send an e-mail when this
recovery completes.
Select this option to specify an e-mail address or addresses to notify upon recovery
completion. If you select this option, you must also enter the e-mail address to notify.
Multiple e-mail addresses must be separated by a comma.
9. On the Summary page, review the recovery settings and then click Recover.
See Also
How to Recover a Storage Group to its Original Location
How to Recover a Database to an Alternate Database
94
How to Copy Exchange Data to a Network Folder
How to Copy Exchange Data to Tape
Recovering Mailboxes
Recovering Data to Clustered Servers
How
to
rec
ov
er
a
dat
ab
as
e
to
an
alt
ern
ate
dat
ab
as
e
1. On the server to which the database will be recovered, configure the target database to
allow it to be overwritten by the recovered data. For instructions, see "How to Configure
the Exchange Databases so That the Restore Process Overwrites Them"
(http://go.microsoft.com/fwlink/?LinkId=97929).
2. In DPM Administrator Console, click Recovery on the navigation bar.
3. Using the browse functionality, select the database to recover.
4. On the calendar, click any date in bold to obtain the recovery points available for that
date. The Recovery time menu lists the time for each available recovery point.
5. On the Recovery time menu, select the recovery point you want to use.
6. In the Actions pane, click Recover.
The Recovery Wizard launches. The wizard options will vary depending on the version of
Exchange.
7. On the Review recovery selection page, click Next.
95
8. Select Recover to another database on an Exchange Server, and then click Next.
9. On the Specify recovery options page, you can select Send an e-mail when this
recovery completes.
Select this option to specify an e-mail address or addresses to notify upon recovery
completion. If you select this option, you must enter the e-mail address to notify. Multiple
e-mail addresses must be separated by a comma.
10. On the Summary page, review the recovery settings, and then click Recover.
See Also
How to Recover a Storage Group to its Original Location
How to Recover a Database to Its Original Location
How to Copy Exchange Data to a Network Folder
How to Copy Exchange Data to Tape
Recovering Mailboxes
Recovering Data to Clustered Servers
96
How
to
co
py
Ex
ch
an
ge
dat
a
to
a
net
wo
rk
fol
der
97
See Also
How to Recover a Storage Group to its Original Location
How to Recover a Database to Its Original Location
How to Recover a Database to an Alternate Database
How to Copy Exchange Data to Tape
Recovering Mailboxes
Recovering Data to Clustered Servers
How
to
co
py
Ex
ch
an
ge
dat
a
to
tap
e
98
copy the data to tape.
• When the data is being copied from tape and the tape library has multiple tape
drives, the library you select in Primary library will read from the source tape and
copy the data to another tape.
• When the data is being copied from tape and the tape library has only a single tape
drive, the library you select in Primary library will read from the source tape and the
library you select in Copy library will copy the data to tape.
9. Enter a label for the tape on which the storage group will be copied.
10. Specify if the data that is copied should be compressed.
11. On the Specify recovery options page, you can select Send an e-mail when this
recovery completes.
Select this option to specify an e-mail address or addresses to notify upon recovery
completion. If you select this option, you must enter the e-mail address to notify. Multiple
e-mail addresses must be separated by a comma.
12. On the Summary page, review the recovery settings, and then click Recover.
Additional Resources
How to Copy a Tape (http://go.microsoft.com/fwlink/?LinkId=196785)
See Also
How to Recover a Storage Group to its Original Location
How to Recover a Database to Its Original Location
How to Recover a Database to an Alternate Database
How to Copy Exchange Data to a Network Folder
Recovering Mailboxes
You can recover deleted e-mail messages using Microsoft Outlook. For instructions, see "How to
Recover a Deleted Item" (http://go.microsoft.com/fwlink/?LinkId=97933). To recover a deleted
mailbox, use the Exchange Management Shell or the Exchange Management Console. For
instructions, see "How to Recover a Deleted Mailbox" (http://go.microsoft.com/fwlink/?
LinkId=97934).
If you cannot recover the mailbox using the Exchange Management Shell or the Exchange
Management Console, such as when the retention period is expired, you can use System Center
Data Protection Manager (DPM) to recover the mailbox.
To recover a mailbox, DPM must copy the entire database because this is the recommended
method that Exchange supports, as explained in Knowledge Base article 904845, "Microsoft
99
support policy for third-party products that modify or extract Exchange database contents"
(http://go.microsoft.com/fwlink/?LinkId=96542).
When you select a mailbox for recovery, you cannot select Latest as the recovery point. The
Latest option recovers the data from the most recent recovery point, and then applies all
committed transactions from the server logs. This functionality is not available for individual
mailboxes.
Item details will not appear on the Recovery Wizard Summary page for Exchange Server
mailboxes.
In This Section
How to Recover an Exchange 2003 Mailbox
How to Recover an Exchange 2007 Mailbox
Ca
uti
on
DPM needs to restore the entire Exchange database before you can recover an
individual mailbox. Please make sure you have sufficient space on the server to which
you are restoring the database.
10
0
How
to
rec
ov
er
a
pre
vio
us
ver
sio
n
of
an
act
ive
Ex
ch
an
ge
20
03
ma
ilb
ox
1. Use the Search tab and a date range to locate the mailbox you want to recover.
2. Select a recovery point for the database that contains the mailbox to be restored.
3. In the Actions pane, click Recover. The Recovery Wizard starts.
4. Review your recovery selection, and then click Next.
5. On the Select Recovery Type page, select Copy to a network folder.
6. On the Specify Destination page, enter a volume on an Exchange server that has a
recovery storage group enabled.
7. On the Select Recovery Options page, select the Bring the database to a clean shut
down state after copying the files check box.
8. Move the database file to the location of the Exchange recovery storage group database.
9. Mount the database under the recovery storage group. For more information, see
Restoring Databases to a Recovery Storage Group in Exchange Server 2003
(http://go.microsoft.com/fwlink/?LinkId=153245).
10. Complete the Recovery Wizard.
11. Extract the mailbox from the recovered database.
• For Exchange Server 2003, use the Microsoft Exchange Server Mailbox Merge
Wizard (ExMerge).
• For Exchange Server 2003 SP1, extract and merge data using Exchange 2003
System Manager.
10
1
How
to
rec
ov
er
a
dis
abl
ed
or
del
ete
d
Ex
ch
an
ge
20
03
ma
ilb
ox
1. Use the Search tab and a date range to locate the mailbox you want to recover.
2. Select a recovery point for the database that contains the mailbox to be restored.
3. In the Actions pane, click Recover. The Recovery Wizard starts.
4. Review your recovery selection, and then click Next.
5. On the Select Recovery Type page, select Recover mailbox to an Exchange server
database.
6. On the Specify Destination page, enter the full names of the Exchange server, including
the domain, storage group, and database.
The database should be dismounted and configured to allow it to be overwritten by the
recovered data. For instructions, see "How to Configure the Exchange Databases so
That the Restore Process Overwrites Them" (http://go.microsoft.com/fwlink/?
LinkId=97929).
7. Complete the Recovery Wizard. DPM recovers the database.
8. Extract the mailbox from the recovered database.
• For Exchange Server 2003, use the Microsoft Exchange Server Mailbox Merge
Wizard (ExMerge).
• For Exchange Server 2003 SP1, extract and merge data using Exchange 2003
System Manager.
See Also
How to Copy Exchange Data to a Network Folder
10
2
Recovering Mailboxes
How to Recover an Exchange 2007 Mailbox
Ca
uti
on
DPM needs to restore the entire Exchange database before you can recover an
individual mailbox. Please make sure you have sufficient space on the server to which
you are restoring the database.
10
3
How
to
rec
ov
er
an
Ex
ch
an
ge
20
07
ma
ilb
ox
for
an
exi
sti
ng
ma
ilb
ox
1. If you do not have an existing Recovery Storage Group, create one by using the new-
storagegroup cmdlet in Exchange Management Shell.
2. Create a recovery database in the Recovery Storage Group by using the new-
mailboxdatabase cmdlet in Exchange Management Shell.
3. Configure the recovery database to allow it to be overwritten by using the set-
mailboxdatabase cmdlet in Exchange Management Shell.
4. In DPM Administrator Console, click the Search tab and select a date range to locate the
mailbox you want to recover.
5. Select a recovery point that contains the mailbox to be restored, and then click Recover.
DPM recovers the database that contains the selected mailbox.
6. On the Review Recovery Selection page, click Next.
7. On the Select Recovery Type page, select Recover mailbox to an Exchange server
database.
8. On the Specify Destination page, enter the full names of the Exchange server, including
the domain, the name of the Recovery Storage Group that you created in step 1, and the
name of the recovery database that you created in step 2.
9. Complete the Recovery Wizard. DPM recovers the database.
10. Configure the destination database to allow it to be overwritten by using the set-
mailboxdatabase cmdlet in Exchange Management Shell.
11. Merge the mailbox data in the recovery database to the production mailbox database,
using the restore-mailbox cmdlet in Exchange Management Shell.
10
4
Example
You need to retrieve some items from a mailbox for an employee who has left the organization.
The following is the identification of the mailbox:
• Exchange Server: exchangeserver1
• Storage group: SG1
• Database: DB11
• Mailbox: John
Storage group SG1 is protected by DPM. You decide to recover the mailbox John to the
manager's mailbox so that he can retrieve the necessary items. The following is the identification
of the manager's mailbox:
• Exchange Server: exchangeserver1
• Storage group: SG2
• Database: DB21
• Mailbox: Simon
To recover the mailbox John to the mailbox Simon, you perform the following steps:
1. Create a Recovery Storage Group (RSG) by running the following Exchange Management
Shell cmdlet:
new-storagegroup -Server exchangeserver1 -LogFolderPath C:\RSG\ -Name RSG
-SystemFolderPath C:\RSG\ -Recovery
This creates a storage group named RSG on exchangeserver1.
2. Add a recovery database to the RSG by running the following Exchange Management Shell
cmdlet:
new-mailboxdatabase -mailboxdatabasetorecover exchangeserver1\SG1\DB11
-storagegroup exchangeserver1\RSG -edbfilepath C:\RSG\DB11.edb
This creates a mailbox on exchangeserver1\RSG\DB11. The .edb file name must be the
same as the .edb file name for the mailbox you are recovering.
3. Set the recovery database to allow overwrites by running the following Exchange
Management Shell cmdlet:
set-mailboxdatabase -identity exchangeserver1\RSG\DB11 -AllowFileRestore 1
4. Open DPM Administrator Console and click Recovery on the navigation bar.
5. Expand the tree and select SG1.
6. Double-click database DB11.
7. Select John, and click Recover.
8. In the Recovery Wizard, on the Review Recovery Selection page, click Next.
9. On the Select Recovery Type page, select Recover mailbox to an Exchange server
database.
10. On the Specify Destination page, enter the following information:
• For Exchange server: exchangeserver1
• For storage group: RSG
10
5
• For database: DB11
11. Specify your recovery options, and then click Recover.
12. Set the destination database to allow overwrites by running the following Exchange
Management Shell cmdlet:
set-mailboxdatabase -identity exchangeserver1\SG2\DB21 -AllowFileRestore 1
The destination database is the database that contains the mailbox to which we want to
recover the e-mail from the John mailbox.
13. When the recovery is complete, run the following Exchange Management Shell cmdlet:
Restore-Mailbox -RSGMailbox 'John' -RSGDatabase 'RSG\DB11' -id 'Simon'
-TargetFolder 'John E-mail'
The manager opens his mailbox and finds a new folder named John E-mail, which contains
the e-mail items from the recovered mailbox.
See Also
Recovering Mailboxes
How to Recover an Exchange 2003 Mailbox
10
6
To
rec
ov
er
the
sto
rag
e
gr
ou
p
or
dat
ab
as
e
to
the
lat
est
poi
nt
in
tim
e
10
7
To
rec
ov
er
the
sto
rag
e
gr
ou
p
to
a
pre
vio
us
poi
nt
in
tim
e
1. Delete the existing log files and checkpoint files on the Exchange server.
2. Set the Exchange server database property Override by restore to True.
3. On the DPM server, recover the storage group or database, selecting the Restore to
original location option.
To recover a storage group or database in clean shutdown state to a network share, you cannot
select Latest as the recovery point.
10
8
To
rec
ov
er
the
sto
rag
e
gr
ou
p
or
dat
ab
as
e
in
cle
an
sh
utd
ow
n
sta
te
to
a
net
wo
rk
sh
are
1. On the DPM server, recover the storage group or database, selecting the Copy to a
network folder option.
2. On the Specify Destination page, specify a folder on a server running Exchange 2007
server.
3. On the Specify Recovery Options page, select the Bring the database to a clean
shutdown state after copying the files option.
4. On the Summary page, click Recover.
10
9
To
rec
ov
er
fro
m
fail
ure
on
the
act
ive
no
de
If the database or logs on the passive node are corrupt, use either of the following procedures to
recover data.
To
rec
ov
er
fro
m
fail
ure
on
the
pa
ssi
ve
no
de
110
5. Remove the common log files (between active and passive nodes) from the passive
node. For example, a failover might have created a new log stream with the same log file
names.
6. On the Exchange server, in Exchange Management Shell, run resume-
storagegroupcopy for the failed storage group.
To
rec
ov
er
fro
m
fail
ure
on
the
pa
ssi
ve
no
de
(if
bot
h
co
pie
s
are
cor
ru
pt)
See Also
Managing Clustered Exchange Servers
Recovering Exchange Data
111
Managing Exchange SCR Servers
System Center Data Protection Manager (DPM) 2010 provides support for backup and recovery
of Microsoft Exchange Server while supporting backup and recovery of the SCR server. For more
information, see Exchange Server 2007 - Standby Continuous Replication
(http://go.microsoft.com/fwlink/?LinkId=134034).
In the current scenario, an onsite DPM server protects the Exchange server or cluster. The DPM
server is protected by an offsite DPM server for disaster recovery. The Exchange server also
replicates its logs and databases to the SCR server at a remote location. By using this
deployment, you make sure that you have disaster recovery options both onsite and offsite.
However, this also means that both Exchange and DPM are sending data across the network.
In the new scenario, instead of having both DPM and Exchange send data over the network, you
use each DPM server to protect the local Exchange servers. The onsite DPM protects the onsite
Exchange server (SCR source); the offsite DPM protects the SCR server (SCR target). This
deployment lets you continue having a disaster recovery scenario both onsite and offsite without
the cost of both applications transporting data over the network.
Depending on business requirements, you can choose to protect either both the SCR source and
the SCR target server or just one of them.
No
te
• Incremental backups are not enabled for all Exchange data sources on an SCR target.
• SCR protection requires a dedicated protection group.
• DPM does not support configurations where a stand-alone Exchange server uses a clustered
SCR or vice versa.
Im
po
rta
nt
If SCR protection was enabled at the time of backup, make sure that it is also enabled at
the time of recovery.
112
Supported Scenarios
• Both SCR source and target servers are Exchange Server 2007 in standalone mode.
• Both SCR source and target servers are Exchange Server 2007 in SCC mode.
• Both CCR source and target servers are Exchange Server 2007 in CCR mode.
In This Section
Protecting an Exchange Server 2007 SCR Target Server Configured as Single Node Cluster
Protecting an Exchange Server 2007 SCR Server in Standalone Mode
Modifying Protection For an Exchange Server 2007 SCR
Recovering an Exchange Server 2007 SCR Server
Stopping Protection for an Exchange Server 2007 SCR Server
Disabling Protection for an Exchange Server 2007 SCR Server
Protecting an Exchange Server 2007 SCR Server Post-Activation
Im
po
rta
nt
If you are using a Single Copy Cluster (SCC) server as your SCR source, you must set
the registry key EnableSccForScr of type DWORD under
HKLM\Software\Microsoft\Microsoft Data Protection Manager\Agent\2.0 to 1 before
proceeding with the following procedure on SCR target server.
113
Pro
ce
du
re
to
en
abl
e
pr
ote
cti
on
for
an
SC
R
ser
ver
1. Create a resource group with same name as the resource group on SCR target cluster.
This is required by the DPM naming convention for clusters. Create an IP address
resource under this resource group and bring it online.
2. Enable SCR protection on the DPM server by using the Enable-ExchangeSCRProtection
script.
You only have to enable SCR protection one time for an SCR server.
Syntax: Enable-ExchangeSCRProtection.ps1 <DPM Server Name>
<Resource Group>.<Cluster FQDN>
Where Cluster FQDN for the cluster must be provided in the format
<Cluster Name>.<Domain>.
Example: Enable-ExchangeSCRProtection.ps1 DPMTest
ExchangeSCRCluster.DRCLUSTER.contoso.com
N
o
t
e
You can check if Exchange SCR protection has been enabled by running get-
ExchangeSCRProtection.ps1.
The value for <Cluster FQDN> must be provided in the following format - <Cluster
Name>.<Domain Name>.
3. Run the Add-SCRSG script from the SCR target server.
114
Syntax: add-SCRSG.ps1 <SCRSourceFQDN> <Storage Group Name>
<Size of Storage Group in MB> <SCRTargetFQDN> <is cluster>
Where target FQDN for the cluster must be provided in the format
<Resource Group Name>.<Cluster Name>.<Domain>.
Example: add-SCRSG.ps1 ExchangeCluster.DRCluster.contoso.com testSG 1024
ExchangeSCRCluster.contoso.com $true
N
o
t
e
N
o
t
e
On the Specify Short-Term Goals page of the Create New Protection Group
wizard, you can only select Express full backups for SCR protection.
I
m
p
o
r
t
a
n
t
115
After activating the SCR server as the primary Exchange server, you must run
Remove-SCRSG.ps1 on the SCR server to enable DPM protection. After
fallback, you must run Add-SCRSG.ps1 on the SCR server to enable DPM
protection.
N
o
t
e
You can check if Exchange SCR protection has been enabled by running get-
ExchangeSCRProtection.ps1.
2. Create a storage group and database, with .log, .sys and .ebd files, with the same name as
the source at some temporary location on the SCR server.
116
C
a
u
t
i
o
n
Ensure that the log file path does not point to the location where the log files for the
SCR source exist. This can lead to the replication service failing.
3. Add the SCR server to the protection group using the Add-SCRSG script from the Exchange
Management Shell.
Syntax: add-SCRSG.ps1 <SCRSourceFQDN> <SGName> <Size of storage group in
MB> <SCRTargetFQDN> <is cluster> <cluster name>
Example: add-SCRSG.ps1 ExchangeServer.contoso.com testSG 1024
ExchangeCluster.contoso.com $false
I
m
p
o
r
t
a
n
t
If the SCR server has been activated (if it is now the primary Exchange server), you
do not need to run this script as the Exchange writer will provide this information to
DPM.
4. Follow the wizard to create a new protection group on the DPM server.
117
Modifying Protection For an Exchange
Server 2007 SCR
The procedure of modifying protection is similar to the procedure to modify protection for an
Exchange server.
No
te
On the Specify Short-Term Goals page of the Modify Group wizard, you can only select
express full backups for SCR protection.
No
te
• On the Recovery tab in DPM Administrator Console, you cannot expand the SCR server to
display individual items.
• You can only recover the SCR server to a network share.
118
Scripts on the SCR server run in the Exchange Management Shell.
No
te
All scripts used in the procedure can be found under the <DPM Installation folder>\Bin
folder of the computer on which the action is being performed. For example, if the script
has to be run on the SCR server, the script is in .\Program Files\Microsoft Data Protection
Manager\DPM\Bin. On the DPM server, the script is in .\Program Files\Microsoft
DPM\DPM\Bin.
Scripts on the SCR server must be run in the Exchange Management Shell.
Syntax (standalone server): Disable-ExchangeSCRProtection.ps1 <DPMServerName>
<ScrPSFQDN>
Example: Disable-ExchangeSCRProtection.ps1 DPMTest ExchangeSCR.contoso.com
Syntax (clustered server): Disable-ExchangeSCRProtection.ps1 <DPMServerName>
<ResourceGroup>.<ClusterFQDN>
Example: Disable-ExchangeSCRProtection.ps1 DPMTest
ExchangeSCRCluster.DRCLUSTER.contoso.com
119
Protecting an Exchange Server 2007 SCR
Server Post-Activation
Activation of an SCR target is done in case of a major disaster that results in no easier options to
recover the Exchange server. When such a disaster happens, you have to remove protection for
the SCR server to allow DPM to protect it as the primary Exchange server.
Use Remove-SCRSG.ps1 on the SCR server to remove it from the protection group.
Im
po
rta
nt
After you activate the SCR server as the primary Exchange server, a new SCR server
has to be put in place to continue SCR protection. You must specifically enable protection
for the new SCR target.
In This Section
Performing General Maintenance on Servers Running SQL
Performing SQL Server Management Tasks
Managing Clustered SQL Servers
Managing Mirrored SQL Servers
Recovering SQL Server Data
No
te
If you disable a protection agent for a server that is a cluster node, you should disable the
protection agent for every node of the cluster.
To
dis
abl
ea
pr
ote
cti
on
ag
ent
In This Section
Using Windows Maintenance Tools on SQL Servers
Performing SQL Maintenance Tasks
Applying Operating System Updates on SQL Servers
Running Antivirus Software on SQL Servers
12
1
Using Windows Maintenance Tools on SQL
Servers
Running Disk Cleanup, Disk Defragmenter, or Chkdsk.exe should have no adverse effect on
performance or data protection.
You should not use other backup applications on a computer running SQL Server that is
protected by DPM.
See Also
Using Windows Maintenance Tools on the DPM Server
Using Windows Maintenance Tools on File Servers and Workstations
Using Windows Maintenance Tools on Exchange Servers
Using Windows Maintenance Tools on SharePoint Servers
Using Windows Maintenance Tools on Virtual Server
12
2
You can use your preferred method for deploying software updates, such as Automatic Updates
or Windows Server Update Services, on computers running SQL Server that are protected by
DPM. Because some software updates require a computer restart, you should schedule or
perform the updates at times that have the least impact on protection operations.
In This Section
Upgrading SQL Server 2000 to SQL Server 2005
Moving SQL Servers Between Domains
How to Rename a Computer Running SQL Server
Changing the Recovery Model of a Database
Replacing a Disk on a SQL Server
Adding Databases to a SQL Server
Changing the Path of a SQL Server Database
Renaming a SQL Server Database
Running Parallel Backups
12
3
3. Add the databases on the upgraded server to a new protection group.
You will be able to use the retained replica to recover data from points in time before the upgrade.
Data created by SQL Server 2000 must be restored to a computer running SQL Server 2000.
You can also use the retained replica to manually create the initial replica for each database in
the new protection group.
No
te
After you reconfigure protection, DPM Administrator Console displays the protected
database as two separate nodes. The protection status in the Protection task area
appears as Inactive replica for one of the database nodes, and the Recovery task area
displays two database nodes with the same name.
12
4
To
ch
an
ge
the
do
ma
in
me
mb
ers
hip
of
a
pr
ote
cte
d
co
mp
ute
r
12
5
• Change the name of a protected computer and associate the existing replicas and recovery
points with the new computer name.
We recommend that you do not change the name of a protected computer. If you must change
the name of a protected computer, you must complete two tasks:
• Remove the data sources on the computer from protection (the old computer name).
• Protect the data source on the computer (the new computer name).
To
ren
am
ea
pr
ote
cte
d
co
mp
ute
r
12
6
the databases, allowing log backups to be
taken. The logs must be truncated explicitly;
otherwise, they continue to grow.
Bulk-logged Similar to the full recovery model except that
certain types of transactions are not logged in
the transaction log.
When a database is added to a protection group, DPM detects the recovery model that the
database is configured to use. DPM does not allow log, or incremental, backups for databases
configured in the simple recovery model. Log backups are only allowed for databases configured
in the full and bulk-logged recovery models.
When the recovery model of a protected database is changed from simple to full or bulk-logged,
DPM protection continues as configured. When the recovery model of a protected database is
changed from full or bulk-logged to simple, express full backups will continue to succeed, but
incremental backups will fail.
To
ch
an
ge
the
rec
ov
ery
mo
del
of
a
pr
ote
cte
d
dat
ab
as
e
to
the
si
mp
le
rec
ov
ery
mo
del
12
7
2. Change the recovery model on the SQL Server database.
3. Add the database to a protection group.
You should also stop protection of a database before you configure log shipping for the database
or change the database to Read Only. After you make the changes to the database, you can
reconfigure protection for the database.
When protecting SQL Server databases that are configured to use the full or bulk-logged recovery
models, DPM creates a folder on the SQL Server that is being protected. This folder is created in
the same location as the first log file (*.ldf) of each protected database.
This folder is used as a temporary store for logs during SQL Server log backup and SQL Server
log restore by DPM. If DPM finds the folder missing, DPM will re-create the folder.
12
8
Changing the Path of a SQL Server Database
When a path associated with a protected database changes, backup jobs will fail. To resolve this
issue, remove the database from protection and then add the database back to the protection
group. This change to the protection group will require a consistency check. After the consistency
check completes successfully, normal protection jobs will resume.
Supported Scenarios
The following is a list of scenarios in which you can perform parallel backups of a SQL data
source.
• The databases are on Microsoft SQL Server 2008.
N
o
t
e
12
9
Unsupported Scenarios
Neither Microsoft SQL Server 2000 nor Microsoft SQL Server 2005 support parallel backups. If
two databases from the same version and instance of SQL Server are scheduled for backup at
the same time, the backup will happen serially.
In This Section
Changing SQL Server Cluster Members
Changing Resource Groups on Clustered SQL Servers
13
0
Changing Resource Groups on Clustered
SQL Servers
A cluster node can have any number of resource groups. Moving a protected data source to a
resource group, between resource groups, or out of a resource group can cause protection job
failures. To successfully make any of those changes to resource group membership, perform the
following steps:
1. Stop existing protection of the data source. The data source could belong to a protection
group as a single data source on a protected server or as a data source as a member of a
resource group.
2. Begin protection of the data source according to its new status, either as a single data source
on a protected server or as a data source as a member of a resource group. This will allocate
a new replica for the data source.
Changing the name of a resource group will affect the protection of all data sources in the
resource group. To change the name of a resource group, perform the following steps:
1. Stop protection of the resource group.
2. Change the name of the resource group.
3. Begin protection of the resource group under its new name.
13
1
Protecting a Mirrored SQL Server Cluster
DPM also supports protection of mirrored SQL Server clusters. The procedure to protect SQL
Server clusters that are mirrored is the same as protecting a mirrored SQL Server database.
No
te
Common Scenarios
No
te
A mirror is broken
When the mirror is broken for a mirrored SQL Server database that is currently protected by
DPM, backups will fail with alerts. Remove protection (with retain data) for the SQL Server
database and reprotect it.
Principal partner in a mirror fails over and fails back before next
backup
This scenario does not affect protection in any manner since DPM is not informed about the fail
over, unless a backup of the mirror is in progress.
13
2
Principal partner fails and the mirror server takes over
DPM will detect that the mirror is now the principal partner, stops the backup job, and performs a
consistency check after 30 minutes on the database that failed over.
No
te
If during these 30 minutes the database fails back to the original principal, DPM will
detect this and resume protection after performing a consistency check.
If you try to backup the mirrored database before the scheduled consistency check (after 30
minutes), an alert indicating that a consistency check is required will be created on the DPM
Monitor tab and the backup will not start until a consistency check is done. To start the backup
immediately, do a consistency check and retry the backup.
Unsupported Scenarios
The following scenarios for mirrored SQL Server databases are not supported by DPM:
• The database is mirrored on the same server.
• The SQL Server mirroring session uses an explicitly-configured IP address.
In This Section
How to Recover a SQL Database to Its Original Location
How to Recover and Rename a SQL Database
How to Recover a Database to a Different Instance of SQL Server
How to Copy a SQL Database to a Network Folder
How to Copy a SQL Database to Tape
How to Recover a SQL Database and Allow Additional Log Backups
13
4
To
rec
ov
er
a
dat
ab
as
e
to
its
ori
gin
al
loc
ati
on
See Also
How to Recover and Rename a SQL Database
How to Recover a Database to a Different Instance of SQL Server
How to Copy a SQL Database to a Network Folder
How to Copy a SQL Database to Tape
How to Recover a SQL Database and Allow Additional Log Backups
13
5
How to Recover and Rename a SQL
Database
To recover and rename a database, use the Recover to any SQL instance option. This option is
unavailable if you select Latest as the recovery point from which to recover the database.
To
rec
ov
er
an
d
ren
am
ea
dat
ab
as
e
See Also
How to Recover a SQL Database to Its Original Location
How to Recover a Database to a Different Instance of SQL Server
How to Copy a SQL Database to a Network Folder
How to Copy a SQL Database to Tape
13
6
How to Recover a SQL Database and Allow Additional Log Backups
No
te
To
rec
ov
er
a
dat
ab
as
e
to
a
diff
ere
nt
ins
tan
ce
of
SQ
L
Se
rve
r
13
7
2. Using either the browse or search functionality, select the database to recover.
3. On the calendar, click any date in bold to obtain the recovery points available for that
date. The Recovery time menu lists the time for each available recovery point.
4. On the Recovery time menu, select the recovery point you want to use. Do not select
Latest for the recovery point.
5. In the Actions pane, click Recover.
The Recovery Wizard starts.
6. On the Review recovery selection page, click Next.
7. Select Recover to any SQL instance, and then click Next.
8. On the Specify recovery destination page, the actions you can take depend on the
version of SQL Server database:
• If you are recovering a database created by SQL Server 2000, specify the alternate
instance of SQL Server to which the database should be recovered. The database
must use the same complete path that it used in its original location.
• If you are recovering a database created by SQL Server 2005, specify the alternate
instance of SQL Server to which the database should be recovered. You can also
specify a path for the database that differs from the path that it used in its original
location.
9. Specify recovery options for network bandwidth usage throttling, SAN-based recovery,
and e-mail notifications, and then click Next.
10. On the Summary page, review the recovery settings, and then click Recover.
See Also
How to Recover a SQL Database to Its Original Location
How to Recover and Rename a SQL Database
How to Copy a SQL Database to a Network Folder
How to Copy a SQL Database to Tape
How to Recover a SQL Database and Allow Additional Log Backups
13
8
To
co
py
a
dat
ab
as
e
to
a
net
wo
rk
fol
der
See Also
How to Recover a SQL Database to Its Original Location
How to Recover and Rename a SQL Database
13
9
How to Recover a Database to a Different Instance of SQL Server
How to Copy a SQL Database to Tape
How to Recover a SQL Database and Allow Additional Log Backups
To
co
py
a
dat
ab
as
e
to
tap
e
14
0
copy the data to another tape.
• When the data is being copied from tape and the tape library has only a single tape
drive, the library you select in Primary library will read from the source tape and the
library you select in Copy library will copy the data to tape.
9. Enter a label for the tape on which the storage group will be copied.
10. Specify if the data that is copied should be compressed or encrypted.
11. On the Set notification page, you can select Send an e-mail when this recovery
completes.
12. On the Summary page, review the recovery settings, and then click Recover.
Additional Resources
How to Copy a Tape (http://go.microsoft.com/fwlink/?LinkId=196785)
See Also
How to Recover a SQL Database to Its Original Location
How to Recover and Rename a SQL Database
How to Recover a Database to a Different Instance of SQL Server
How to Copy a SQL Database to a Network Folder
How to Recover a SQL Database and Allow Additional Log Backups
14
1
To
rec
ov
er
a
dat
ab
as
e
wit
ho
ut
tra
ns
act
ion
roll
ba
ck
14
2
the log backup sequences to be correctly ordered.
14. Use the Restore command with the Log argument to apply the desired logs to the
database in the right order.
For more information on the Restore command, see RESTORE Arguments (Transact-
SQL) (http://go.microsoft.com/fwlink/?LinkId=104665).
See Also
How to Recover a SQL Database to Its Original Location
How to Recover and Rename a SQL Database
How to Recover a Database to a Different Instance of SQL Server
How to Copy a SQL Database to a Network Folder
How to Copy a SQL Database to Tape
In This Section
Configuring SharePoint Protection
Protecting a SharePoint Farm
Protecting SharePoint Front-End Web Server
Protecting SharePoint Search
Recovering SharePoint Data
Performing SharePoint Protection Management Tasks
Performing General Maintenance on Servers Running SharePoint
Troubleshooting SharePoint Protection and Recovery
In This Section
Configuring the DPM Server for SharePoint Protection
Configuring SharePoint Farm Servers
In This Section
Configuring Data Source Groups
14
4
Configuring DataSourceGroups.xml
To configure the DatasourceGroupsSample.xml file, follow these steps:
1. Locate the DatasourceGroupsSample.xml file in the Config folder under the DPM installation
folder.
2. Make a copy of the DatasourceGroupsSample.xml file and rename it.
After you perform the preceding steps, when a new protection group is created or an existing
protection group is modified, DPM will load the DatasourceGroupsSample.xml file.
No
te
Modifying an existing protection group leads to cancellation of all running jobs. To avoid
this, you can create a dummy protection group that contains a single data source and
modify it when you want to reload the XML file.
The DatasourceGroupsSample.xml file will contain a single group (for example, Group1) and all
the data sources protected by the DPM server will be present in that group. If you do not want to
control the backup of any particular data source, then that data source can be removed from the
XML file. The following is an example of a DatasourceGroupsSample.xml file.
No
te
The following table describes the field names in the DatasourceGroupsSample.xml file.
Name Description
DataSourceName Name of the database. The name should be in
the following format:
SQLInstanceName\SQLDBName.
ProtectedServerName Name of the server where the database
resides.
WriterID ID of the VSS writer that reports this database.
This field is set by the system and does not
have to be modified.
14
5
When running parallel backups, DPM selects only one database from each group.
If you have to back up two databases from the same spindle or disk, ensure that both the
databases are in the same group in the XML file. This helps to maximize the use of disk
throughput and also ensures that DPM does not back up both the databases in parallel.
You can create an unlimited number of groups. However, DPM can run backups for only eight
SharePoint databases from the same database instance in parallel. In SQL Server 2005, you can
run only one backup at a time from a given instance of SQL Server.
Unloading DatasourceGroups.xml
To unload the datasourcegroups.xml file, perform the following steps:
1. Save a copy of datasourcegroups.xml using a different name.
2. Empty datasourcegroups.xml and save it.
3. Run the Modify Protection Wizard for any protection group to complete the unload process.
In This Section
Configuring the Front-End Web Server
Configuring the SQL Backend Servers
14
6
Prerequisites for Front-end Web Server
If you are running Windows Server 2003 on the front-end Web server and have installed
Knowledge Base article 940349, then you must ensure that the following prerequisites are
installed:
• Windows SharePoint Services 3.0 with SP2 or Windows SharePoint Services 3.0 with SP1 or
Microsoft Office SharePoint Server 2007 with SP2 or
Microsoft Office SharePoint Server 2007 with SP1 or Microsoft SharePoint Foundation 2010
or Microsoft SharePoint Server 2010.
• At least 2 GB of space on the volume in which the DPM is installed for every 10 million items
in the farm. This is required for catalog generation. In DPM, to perform granular level
recovery of items (site collections, sites, lists, document libraries, folders, individual
documents and list items), catalog generation provides you a list of URLs contained within
each content database. This list of URLs is displayed on the recoverable item pane in the
DPM Administrator console.
• If you are running Windows SharePoint Services 3.0 SP1 or MOSS 2007 SP1, then install
Knowledge Base article 941422, Update for Windows SharePoint Services 3.0
(http://go.microsoft.com/fwlink/?LinkId=100392).
N
o
t
e
You must install Knowledge Base article 941422 on all protected servers on which
Windows SharePoint Services 3.0, Microsoft Office SharePoint Server 2007 with
SP1, and Microsoft Office SharePoint Server 2007 are installed.
• Run ConfigureSharePoint.exe on the front-end Web server. For more information about using
ConfigureSharePoint, see Using ConfigureSharePoint.
14
7
N
o
t
e
If the SharePoint VSS Writer service does not start on the front-end Web server then
you must manually start this service. For more information, see Starting and
Configuring the WSS VSS Writer Service in Configuring DPM 2010
(http://go.microsoft.com/fwlink/?LinkId=179157).
• In the SharePoint farm, if you have SQL Server databases that are configured with SQL
Server aliases, then install the SQL Server client components on the front-end Web server
that DPM will protect. For information about installing SQL Server 2005 components, see
How to: Install SQL Server 2005 (http://go.microsoft.com/fwlink/?LinkId=147257).
Using ConfigureSharePoint
Before you begin to protect a SharePoint farm, you must configure protection for SharePoint by
using the ConfigureSharePoint.exe tool.
In DPM, ConfigureSharePoint.exe is a tool that is required to be run on the front-end Web server
from where you plan to protect SharePoint farm data. The ConfigureSharePoint.exe file can be
found in the <DPM Installation Path>\bin folder on the front-end Web server. This tool must be run
in the following scenarios:
• Before you begin to protect a SharePoint farm
• Change in SharePoint farm administrator password
• Change in SharePoint farm administrator account
Permissions
To run the ConfigureSharePoint.exe tool, ensure that you meet the following prerequisites:
You must be a member of the Administrators group on the local computer to run this tool.
You must run this tool from an elevated command prompt.
The ConfigureSharePoint.exe tool provides the following permissions to the farm administrator on
the front-end Web server:
• Read and Execute to all DPM directories: DPM has to load the DLLs from the DPM Bin
directory when WSSCmdletWrapper.exe runs.
14
8
• Read, Execute, and Write (all) access on the Temp directory in the DPM directory: DPM has
to create a directory inside the DPM Temp directory where item-level catalog dumps are
created. DPM also creates a log file, WSSCmdletWrapperCurr.errlog, inside the DPM Temp
directory.
• Read permissions to the DPM hive in the registry.
Syntax
ConfigureSharePoint [-EnableSharePointProtection] [-EnableSPSearchProtection] [-
ResolveAllSQLAliases] [-SetTempPath <path>]
No
te
To run this command, you must be a local administrator on the front-end Web server. In
Windows Server 2008 and later versions, ensure that you run this command from an
elevated command prompt.
Parameters
Parameter Description
EnableSharePointProtection • To enable SharePoint farm protection using
a DPM server, you must ensure that you
run this option on the front-end Web server
from where you plan to protect SharePoint
farm data.
Do not run this option on more than one
front-end Web server.
To run this option on multiple servers, run
the command “Stsadm –o
unregisterwsswriter” on the front-end Web
servers from where you do not plan to
protect SharePoint farm data.
This option performs the following:
• Enables the SharePoint VSS writer required
for SharePoint farm protection.
• Registers the identity of the DCOM
application WssCmdletsWrapper to run as
a user whose credentials are entered with
this option. If you are prompted to enter
your user credentials, then enter the
14
9
credentials of a farm administrator.
EnableSPSearchProtection • You must run this option from any one of
the front-end Web servers from where you
plan to protect the Windows SharePoint
Services 3.0/MOSS 2007 Search service.
This server can be an indexing service or
any other front-end Web server.
Do not run this option on multiple servers.
If you want to run this option on multiple
servers, then delete the registry key
SharePointSearchEnumerationEnabled
under HKLM\Software\Microsoft\
Microsoft Data Protection
Manager\Agent\2.0\ on the front-end Web
server that is not used for protecting
SharePoint Search services.
• Enables the protection of SP Search and
MOSS 2007 SSP by using the registry key
SharePointSearchEnumerationEnabled
under HKLM\Software\Microsoft\
Microsoft Data Protection
Manager\Agent\2.0\ on the front-end Web
Server.
• Registers the identity of the DCOM
application WssCmdletsWrapper to run as
a user whose credentials are entered with
this option. If you are prompted to enter
your user credentials, then enter the
credentials of a farm administrator.
ResolveAllSQLAliases This option displays all the aliases reported by
the SharePoint VSS writer and resolves them to
the corresponding SQL Server. It also displays
their resolved instance names. If the servers
are mirrored, it also displays the mirrored
server. It reports all the SQL Server aliases that
are not being resolved to a SQL Server.
No
te
15
0
run the ConfigureSharePoint [-
EnableSharePointProtection] or
ConfigureSharePoint [-
EnableSPSearchProtection] command
on the front-end Web server.
SetTempPath Sets the environment variables TEMP and TMP
to the specified path. Item-level recovery fails if
a large site collection, site, list, or item is being
recovered and there is insufficient space in the
farm administrator Temporary folder. This option
allows you to change the folder path of the
temporary files to a volume that has sufficient
space to store the site collection or site being
recovered.
No
te
DPM supports Standard, Enterprise, Workgroup, and Express Editions of SQL Server.
You must start the SQL Server VSS Writer Service on computers running SQL Server before you
can start protecting SQL Server data. The SQL Server VSS Writer Service is turned on by default
on computers running SQL Server. To start the SQL Server VSS Writer service, in the Services
console, right-click SQL Server VSS writer, and then click Start.
15
1
1. Install DPM protection agents on all SharePoint servers that you plan to protect. For more
information about how to install DPM agents, see Installing Protection Agents
(http://go.microsoft.com/fwlink/?LinkID=179676).
2. Enable SharePoint protection by configuring a front-end Web server. For more information
about configuring front-end Web servers, see Configuring the Front-End Web Server.
3. If you are protecting Windows SharePoint Services 3.0/MOSS 2007 Search, you must
configure a front-end Web server to enable SharePoint Search protection.
4. On the DPM server, run the Create New Protection Group Wizard to protect SharePoint data
that exists under a front-end Web server that is configured for protection. For more
information about how to create a protection group, see DPM Help.
In This Section
Protecting a SharePoint Farm by Using Mirrored Databases
Protecting a SharePoint Farm by Using Databases With SQL Server Aliases
Long-Term Protection for a SharePoint Farm on Tape
Prerequisites
• Install the DPM protection agent on both the computers that are running the instance of SQL
Server and hosts the principal and mirror database.
No
te
DPM does not support mirroring the database on the same instance of SQL Server.
15
2
Common Scenarios
A protected SharePoint database gets mirrored
At the time of the backup, DPM detects that the database has been mirrored and raises an alert
that farm configuration has been changed. DPM treats the mirrored database as a new database
in the farm and automatically protects it. Although the alert is not deactivated, all the content in
the SharePoint farm will continuously be protected. To inactivate this alert, you must stop
protection for the farm (with retain data) and re-protect it.
No
te
Even if failovers between principal and mirrored copies of the database occur, DPM
maintains a single replica for the mirror.
Principal partner in a mirror fails over and fails back before next
backup
This scenario does not affect protection in any way unless a backup is in progress.
No
te
If, during these 30 minutes, the database fails back to the original principal, DPM detects
this and resumes protection after performing a consistency check.
A mirror is broken
When the mirror is broken for a mirrored SQL Server database that is currently protected by
DPM, backups fail with alerts. This is similar to the behavior when the protected SharePoint
database gets mirrored.
15
3
Protecting a SharePoint Farm by Using
Databases With SQL Server Aliases
System Center Data Protection Manager (DPM) 2010 supports protection of a SharePoint farm
that uses SQL Server databases configured with SQL Server aliases. This additional support
does not require any major changes in the procedure you use to protect or recover SharePoint
data with DPM.
Im
po
rta
nt
Prerequisites
• You must configure SQL Server aliases and define all SQL Server aliases in the farm on the
front-end Web Server. We recommend that you use one SQL Server alias per database.
• The SQL Server client connectivity components must be installed on the front-end Web
Server.
Additional Resources
How to: Create a Server Alias for Use by a Client (SQL Server Configuration Manager)
(http://go.microsoft.com/fwlink/?LinkId=132909)
15
4
Checks the latest recovery point on the disk and copies the databases on the disk to the tape.
Checks the previous recovery point for all the databases that were missing in the latest recovery
point.
Similarly, DPM traverses older recovery points one-by-one unless the recovery points of all the
databases are found.
If there is a database for which a recovery point was not created since the last successful
scheduled tape backup, then that database’s recovery point on the tape fails.
No
te
Long-term protection for a SharePoint farm on tape is available only on the primary DPM
server.
15
5
Protecting SharePoint Search
The procedure to protect a SharePoint Search requires a few tasks to be performed before you
start the Create New Protection Group Wizard. For details on protecting Windows SharePoint
Services 3.0 SP Search and MOSS 2007 Shared Services Provider (SSP), see the other topics in
this section.
In This Section
Protecting Windows SharePoint Services 3.0 SP Search Service Data
Protecting MOSS 2007 Shared Services Provider (SSP) Search
15
6
Protecting MOSS 2007 Shared Services
Provider (SSP) Search
With System Center Data Protection Manager (DPM) 2010, you can backup and restore
SharePoint Search created by Microsoft Office SharePoint Server 2007 (MOSS).
To
Pr
ote
ct
a
Co
mp
ute
r
tha
t is
Ru
nni
ng
M
OS
S2
00
7
SS
P
15
7
No
te
During the backup process, DPM pauses index crawling and all background processes.
After the backup process is complete, DPM automatically resumes these processes. This
does not affect the search function.
C
a
u
t
i
o
n
In This Section
Recovering SharePoint Front-End Web Server
Recovering SharePoint Farm Content
Recovering SharePoint Web Application
Recovering SharePoint Content Database
Recovering SharePoint Content Database
15
8
Recovering SharePoint Items
Recovering SharePoint Search
15
9
Ca
uti
on
To
rec
ov
er
far
m
dat
a
to
a
fun
cti
oni
ng
far
m
16
0
To
rec
ov
er
far
m
dat
a
wh
en
the
pr
ote
cte
d
far
m
is
un
av
ail
abl
e
1. Create a new farm that uses the same instance of SQL Server and the same front-end
Web server as the original protected farm.
2. On the front-end Web server that DPM uses to recover farm data, run the following
command at the command prompt::
ConfigureSharePoint-EnableSharePointProtection
3. On the DPM server, in DPM Administrator Console, click Recovery on the navigation bar.
4. In the Protected data pane, expand the server that contains the farm you want to
recover, and then click All Protected SharePoint Data.
The farm displays in the Recoverable item pane as server name\farm name.
5. Use the calendar and the Recovery time menu to select a recovery point.
6. In the Recoverable item pane, click the farm item.
7. In the Actions pane, click Recover.
8. Complete the wizard.
9. On the main front-end Web server for the server farm, run the SharePoint Products and
Technologies Configuration Wizard and disconnect the front-end Web server from the
farm.
16
1
N
o
t
e
If the main front-end Web server for the server farm is not the front-end Web
server that DPM uses to protect the farm, you must also disconnect the front-end
Web server that DPM uses to protect the farm.
10. Open Internet Information Services (IIS) and delete all Web site and application pool
entries related to the farm.
11. Run the SharePoint Products and Technologies Configuration Wizard, select to connect
to an existing server farm, and specify the server name and database name for the farm
you created in step 1.
N
o
t
e
Perform step 11 for all front-end Web servers for the server farm.
12. On the Completing the SharePoint Products and Technologies Configuration
Wizard page, click Advanced Settings, and then click Next.
13. On the Advanced Settings page, select the option Use this machine to host the web
site, and complete the wizard.
In This Section
Recovering a SharePoint Farm by Using Databases with SQL Server Aliases
Recovering a SharePoint Farm by Using Mirrored Databases
16
2
Recovering a SharePoint Farm by Using
Databases with SQL Server Aliases
See Also
Recovering SharePoint Data
Name Description
Typical recovery Select this option to recover the mirrored SQL
Server databases to the instances of SQL
Server that were hosting these databases as
the principal database when the selected
16
3
recovery point was created.
Custom recovery Select this option to recover the mirrored SQL
Server databases of the SharePoint farm to the
instances of SQL Server. At the point of
recovery point creation, for each mirrored SQL
Server database, you can select either of its
partner instances of SQL Server
(principal/mirror). Before you select the
instance of SQL Server, make sure of the
following:
• The selected instance of SQL Server is
online.
• The SQL alias being used on the front-end
Web server points to the selected instance
of SQL Server.
The default selection is the partner from which the database was last backed up.
If you are using a SQL Server alias for the mirrored database, then before the recovery ensure
that aliases corresponding to the respective databases are configured such that they refer to the
SQL Server instance location selected on the Recovery Wizard page.
Otherwise the farm recovery fails at the end because it cannot attach the databases after
recovery. For more information, see Recovering a SharePoint Farm by Using Databases with
SQL Server Aliases.
See Also
Recovering SharePoint Data
16
4
Pro
ce
du
re
to
rec
ov
er
a
We
b
ap
pli
cat
ion
N
o
t
e
If you want to retain the same URL for the application, ensure that the host
header and port are the same as the original application.
2. Restore all the databases in that Web application to either the original location or to a
different SQL Server.
3. Attach each content database to the Web application by using the stsadm command or
from the Central Administrator site
Additional Resources
Addcontentdb: Stsadm operation (Office SharePoint Server) (http://go.microsoft.com/fwlink/?
LinkId=143331)
16
5
No
te
No
te
To recover a mirrored database to its original instance of SQL Server, select the Recover
to any SQL Instance option, and then, on the Specify Alternate Recovery Location
page, specify the recovery destination path of the original instance of SQL Server.
In This Section
Using a Recovery Farm
Recovering a SharePoint Site Collection
Recovering a SharePoint Site
Recovering a List, List Item, Document Library, or Document
DPM Cataloging to Recover SharePoint Items
16
6
Using a Recovery Farm
A recovery farm is a temporary staging SharePoint where the content database that contains the
item to be restored is temporarily hosted. The SharePoint APIs extract the item from the content
database in the recovery farm and then import the item into the target farm.
In This Section
Creating a Recovery Farm
No
te
The version of SQL Server must be the same or higher than what was installed at the
time of backup.
SQL Server VSS Writer should be running on the recovery farm.
This computer must be separate from the DPM server, Active Directory, domain
controller, any server on which SharePoint data is protected by DPM and farm
computers.
To
cre
ate
a
rec
ov
ery
far
m
1. Install the DPM agent on the recovery farm computer through the DPM server backing up
the farm.
2. Run ConfigureSharepoint.exe. For more information on running
ConfigureSharePoint.exe, see Using ConfigureSharePoint.
16
7
3. If you protect a MOSS farm, then the recovery farm must also be MOSS.
4. The features and templates installed on the recovery farm must match those of the target
farm as it was at the time of backup. Any customized templates, added or modified, on
the production farm, must be added to the recovery farm to ensure a successful recovery.
N
o
t
e
You can enable all the features and templates installed on the recovery farm and
use it for the different farms existing in your SharePoint environment.
5. If a service pack or update is installed on the protected farm, the recovery farm must
have the same service pack or update installed otherwise item-level restore operations
could fail.
6. Both the recovery and target farms must be in the same language and have the same
language packs installed.
7. Create a Web application and name it DPMRecoveryWebApplication. To create a new
Web application, see the instructions at Create or extend Web applications (Windows
SharePoint Services) (http://go.microsoft.com/fwlink/?LinkId=94374).
8. Ensure that no content database is already attached to the recovery web application
(DPMRecoveryWebApplication) as this will cause recoveries to fail. The web application
name DPMRecoveryWebApplication is a required name and it must be created for
DPM to be able to restore any SharePoint data.
When you restore a site, DPM restores the database to the recovery farm, extracts the site
from the recovery farm, and imports it into the target farm. During this process, DPM creates
a temporary file on the recovery farm at a location specified in the Recovery Wizard. You
should periodically delete the temporary files at that location.
No
te
The recovery farm must have enough hard disk space to store the largest content
database in the environment. Best practice would dictate that an additional 10-20%
be allocated on the temporary storage volume to provide a cushion for growth and
reduce the risk of running out of space when trying to recover time-sensitive
SharePoint data.
16
8
See Also
Create a recovery farm (Office SharePoint Server 2007)
Pro
ce
du
re
to
rec
ov
er
a
sit
e
col
lec
tio
n
1. Create an empty site collection with the same name on the target farm.
2. Apply the same features and templates to the site collection that was used at the time of
backup. If they do not match, SharePoint will raise an error to indicate that the site
templates do not match, which causes the DPM site restoration to fail.
3. Recover the top-level site from the site collection. For more information about how to
recover a SharePoint site, see Recovering a SharePoint Site.
N
o
t
e
If a content database contains only one site collection, you can chose to recover
the database directly, and then attach it to the farm using the stsadm command.
16
9
Additional Resources
Addcontentdb: Stsadm operation (Office SharePoint Server) (http://go.microsoft.com/fwlink/?
LinkId=143331)
No
te
This recovery process does not work with lists. If you are trying to restore lists, you must
manually delete all lists under the site, and then restore the site.
Whether you recover a SharePoint site to its original location or to another location on the same
farm, the overall steps are the same. First, you create a recovery farm, and then you use DPM to
recover the site by using the recovery farm.
No
te
You can use DPM to recover items (site collections, sites, document libraries, lists,
documents, and list items) from a Microsoft SharePoint Foundation 2010 or Microsoft
Office SharePoint Server 2010 farm both with and without a recovery farm.
17
0
To
rec
ov
er
a
Sh
are
Poi
nt
sit
e
1. Create a farm that DPM can use for the recovery. To create a recovery farm, see the
instructions at Creating a Recovery Farm.
2. In DPM Administrator Console, click Recovery on the navigation bar.
3. In the Protected data pane, expand the server that contains the farm you want to
recover, and then click All Protected SharePoint Data.
The farm appears in the Recoverable Item pane as server name\farm name.
4. Double-click the farm item.
The databases for the farm appear in the Recoverable Item pane.
5. Navigate to the recoverable item objects and locate the site that you want to recover.
6. Select a recovery point for the site that you want to recover, and then, in the Actions
pane, click Recover.
7. On the Review Recovery Selection page, confirm that the correct item is being
recovered based on Recovery Item.
8. On the Select Recovery Type page, select one of the following options:
• Recover to original site
• Recover to an alternate site
9. This step applies only to SharePoint 2010:
On the Select Recovery Process page, select one of the following options:
a. Recover without using a recovery farm. Select this option if the version of the
target Microsoft SharePoint 2010 farm is the same as the version at the time of the
selected recovery point, and then click Next.
i. On the Specify Temporary Server page, do the following:
• In the SQL instance field, browse to the instance of SQL Server that can be
used temporarily to stage a copy of the SharePoint content database.
The temporary instance of SQL Server can be:
17
1
the front-end Web server of the protected SharePoint farm
noteDXDOC112778PADS Note
If you are using DPM’s instance of SQL Server or any other instance of SQL Server, ensure
that its version is equal to or later than the version of the production SQL Server. The
selected instance of SQL Server can be a Microsoft Cluster Server (MSCS).
• In the Database file location field, browse to the instance of SQL Server on
that server, and then select the temporary location where the database files
can be copied.
b. Recover using a recovery farm. Select this option if the version of the target
Microsoft SharePoint 2010 farm has changed since the selected recovery point was
created, and then click Next. For more information about how to create a recovery
farm, see Creating a Recovery Farm.
i. On the Specify Temporary Server page, do the following:
1. In the Front-end Web server field, browse for the recovery farm server
where DPMRecoveryWebApplication has been created to temporarily
stage data prior to recovery.
2. In the SQL instance field, browse to the instance of SQL Server that can
be used temporarily to stage a copy of the SharePoint content database
that contains the requested site before recovery.
3. In the Database file location field, browse to the instance of SQL Server
on that server and then select the temporary location where the database
files can be copied to a recovery farm.
10. This step applies only if you are recovering to an alternate location.
In the Recovery target site section, enter the URL for the alternate site. A site can be
restored to a different location within the same farm to which it belongs. Therefore,
specify a URL within the same SharePoint farm under which you want to recover the
selected SharePoint site.
17
2
N
o
t
e
The target site URL must be based on the same site template as the site that is
being restored. For example, SharePoint will not allow a site that was created by
using a Wiki Site template to be restored onto a site that was created by using a
Team Site, Blank Site, Blog, or Document Workspace templates. A custom
template must reside on the recovery farm and be used to create the alternate
site to which the recovery is being made.
11. On the Specify Staging Location page, enter a directory where the SharePoint can be
temporarily stored pending recovery to the original or alternate site.
N
o
t
e
17
3
I
m
p
o
r
t
a
n
t
This is an important consideration if there have been any changes to the security
settings since the recovery point was taken.
13. On the Summary page, confirm all settings, and then click Recover to begin the
recovery process.
See Also
Recovering SharePoint Data
Recovering SharePoint Items
17
4
To recover an item to its original location
N
o
t
e
You can only select and recover one object at a time. If you want to recover more
than one object, consider recovering a higher level folder to an alternate location
and then recovering the individual objects from within the SharePoint Central
Administration website.
5. Click Recover in the Actions pane and confirm the recovery details on the Review
Recovery Selection page.
6. On the Select Recovery Type page, select Recover to original site.
7. The following steps apply to SharePoint 2010:
On the Select Recovery Process page select any one of the following two options that
are listed below:
a. Recover without using a recovery farm. Select this option if the version of the
target Microsoft SharePoint 2010 farm is same as at the time of the selected recovery
point. Click Next
i. On the Specify Temporary Server page
1. In the SQL instance field browse for the instance of SQL Server that can
be used temporarily to stage a copy of the SharePoint content database
that contains the requested item before recovery.
2. In the Database file location field, browse for the instance of SQL
17
5
Server on that server and then select the temporary location where the
database files can be copied
The temporary instance of SQL Server can be:
noteDXDOC112778PADS Note
If you are using DPM’s instance of SQL Server or any other instances of SQL Server then
make sure that its version is equal to or a has a later version than the version of the
production SQL Server. The selected instance of SQL Server can be a Microsoft Cluster
Server (MSCS).
b. Recover using a recovery farm. Select this option if the version of the target
Microsoft SharePoint 2010 farm has changed from the time, the selected recovery
point was created. Click Next
i. On the Specify Temporary Server page, enter the information for recovery farm.
For more information about how to create a recovery farm, see Creating a
Recovery Farm.
1. In the Front-end Web server field, browse for the recovery farm server
where DPMRecoveryWebApplication has been created to temporarily
stage data prior to recovery.
2. In the SQL instance field, browse for the instance of SQL Server that
can be used temporarily to stage a copy of the SharePoint content
database that contains the requested item before recovery.
3. In the Database file location field, browse for the instance of SQL
Server on that server and then select the temporary location where the
database files can be copied on to a recovery farm.
8. On the Specify Staging Location page, enter a directory where the SharePoint data will
be temporarily stored, pending recovery to the original site.
9. On the Specify Recovery Options page, specify whether the recovery point’s security
settings or the original site’s security settings will be applied to the recovered data object
in the Restore Security section.
This is an important consideration if there have been security settings changes since the
recovery point was taken.
17
6
N
o
t
e
The network bandwidth usage throttling is used when there are concerns about
the restore process consuming excessive bandwidth.
The SAN Recovery option is only available if the attached SAN is capable of
snapping clones and splitting clones.
The Notification section is simply to notify administrators and other personnel of
the completion of the recovery process.
10. Confirm the settings on the Summary page and click Recover to begin the process.
No
te
An alternate location can be on the same SharePoint farm and merely using a different
site name or port number.
1. Create a farm that DPM can use for the recovery. For more information, go Creating a
Recovery Farm.
2. In DPM Administrator Console, click Recovery on the Actions pane.
3. In the Protected data pane, expand the server that contains the farm you want to
recover, double-click All Protected SharePoint Data, and then double-click the server
farm name.
Content databases display in the Recoverable item pane.
4. Use the calendar and Recovery time menu to select a recovery point.
5. In the Recoverable item pane, select the content database and browse to the item you
wish to recover.
17
7
N
o
t
e
You can only select and recover one object at a time. If you want to recover more
than one object, consider recovering a higher level folder to an alternate location
and then recovering the individual objects from within the SharePoint Central
Administration website.
6. Click Recover in the Actions pane and confirm the recovery details in the Review
Recovery Selection page.
7. On the Select Recovery Type page, select Recover to an alternate site.
8. The following steps apply to SharePoint 2010:
On the Select Recovery Process page select any one of the following two options that
are listed below:
a. Recover without using a recovery farm. Select this option if the version of the
target Microsoft SharePoint 2010 farm is same as at the time of the selected recovery
point. Click Next
i. On the Specify Temporary Server page
1. In the SQL instance field browse for the instance of SQL Server that can
be used temporarily to stage a copy of the SharePoint content database
that contains the requested item before recovery.
2. In the Database file location field, browse for the instance of SQL
Server on that server and then select the temporary location where the
database files can be copied
The temporary instance of SQL Server can be:
noteDXDOC112778PADS Note
If you are using DPM’s instance of SQL Server or any other instances of SQL Server then
make sure that its version is equal to or a has a later version than the version of the
17
8
production SQL Server. The selected instance of SQL Server can be a Microsoft Cluster
Server (MSCS).
b. Recover using a recovery farm. Select this option if the version of the target
Microsoft SharePoint 2010 farm has changed from the time, the selected recovery
point was created. Click Next
i. On the Specify Temporary Server page, enter the information for recovery farm.
For more information about how to create a recovery farm, see Creating a
Recovery Farm.
1. In the Front-end Web server field, browse for the recovery farm server
where DPMRecoveryWebApplication has been created to temporarily
stage data prior to recovery.
2. In the SQL instance field, browse for the instance of SQL Server that
can be used temporarily to stage a copy of the SharePoint content
database that contains the requested item before recovery.
3. In the Database file location field, browse for the instance of SQL
Server on that server and then select the temporary location where the
database files can be copied on to a recovery farm.
9. In the Recovery target site field, enter the URL for alternate site. An item can be restored
to a different location within the same farm to which it belongs to. Therefore specify a
URL within the same SharePoint farm under which you would want to recover the
selected SharePoint item.
N
o
t
e
The site URL entered into the Target site URL field must be based on the same
site template as the site hosting the object which is being restored. For example,
SharePoint will not allow an object created in a site using a ‘Wiki Site’ template to
be restored onto a site created using the ‘Team Site’, ‘Blank Site’, ‘Blog’, or
‘Document Workspace’ templates. If custom templates have been used, those
same templates must reside on the recovery farm as well as having been used to
create the alternate site where the recovery is being made to.
10. On the Specify Staging Location page, enter a directory where the SharePoint data will
be temporarily stored, pending recovery to the original site.
11. On the Specify Recovery Options page, specify whether the recovery point’s security
settings for the object being recovered or the original site’s security settings will be
applied to the recovered data object in the Restore Security section. This is an important
17
9
consideration if there have been security settings changes since the recovery point was
taken.
N
o
t
e
The network bandwidth usage throttling is used when there are concerns about
the restore process consuming excessive bandwidth from bandwidth sensitive
applications.
The SAN Recovery option is only available if the attached SAN is capable of
snapping clones and splitting clones.
The Notification section is simply to notify administrators and other personnel of
the completion of the recovery process.
12. Confirm the settings on the Summary page and click Recover to begin the process.
No
te
See Also
Recovering SharePoint Data
Recovering a SharePoint Site
18
0
By default, the catalog task is scheduled to run three hours after the first scheduled backup of the
corresponding SharePoint farm in the day. To modify the default schedule, run the Set-
ProtectionJobStartTime cmdlet in DPM Management Shell on the DPM server.
In DPM 2010, creating a catalog of the farm (list of URLs within the farm) is tightly tied to the
backup of the SharePoint farm.
Syntax
Set-ProtectionJobStartTime–ProtectionGroup <ProtectionGroup Object> –CatalogOffset
<Offset in Minutes>
Parameters
Parameter Description
ProtectionGroup Provide the protection group that contains the
SharePoint farm.
No
te
When you change the offset for the catalog task, ensure that cataloging for the SharePoint farm
begins only after the recovery point prior to the task is completed.
In This Section
Recovering Windows SharePoint Services 3.0 SP Search Service Data
Recovering MOSS 2007 Shared Services Provider (SSP) Search
Im
po
rta
nt
The farm administrator should have administrator rights on the Indexing Service. If this is
not the case, DPM is not able to stop the search service to ensure a proper recovery.
Recovering to original location
18
2
The procedure to recover SP Search data is similar to the recovery of any data source. You begin
the recovery process from the DPM Administrator Console. This brings up the Recovery wizard
which guides you through the process. The recovery process automatically deletes existing index
files and resumes the SP Search service after the recovery is complete.
In the case of disaster recovery, you must configure SP Search with the original configuration of
the latest recovery point before performing a recovery.
Rec
ov
eri
ng
as
file
s
18
3
Recovering MOSS 2007 Shared Services
Provider (SSP) Search
I
m
p
o
r
t
a
n
t
If you are recovering to the original location, you must delete the SSP and its
index files from the original location before proceeding with recovery. This must
be done even if it is the default SSP using the -Force parameter. The location
details of these files are available on the Summary page of the Recovery wizard.
3. After the recovery process is completed, you must run RestoreSSP on the protected
computer with the KeepIndex parameter to ensure that the index file is not reset during
the process of recreating the SSP.
18
4
How to Recover an Index with Mirrored Database
in MOSS 2007 Shared Services Provider (SSP)
Search
In the case where the MOSS farm uses a mirrored database, you cannot recover the index to the
original location. In this case, you must recover the individual components of the index and
manually reattach them to the instance of SQL Server.
Additional Resources
Restoressp: Stsadm operation (Office SharePoint Server) (http://go.microsoft.com/fwlink/?
LinkId=113889)
In This Section
Changing the SharePoint Farm Administrator Password
Adding a Database to a SharePoint Farm
Removing a Database from a SharePoint Farm
Adding or Removing Servers in SharePoint Farm
Switching the Front-End Web Server
Upgrading SharePoint
Moving SharePoint Servers Between Domains
18
5
Renaming SharePoint Server
Improving DPM Recovery Search for SharePoint Items
18
6
The following recommended actions associated with this alert are provided in the alert details:
1. In DPM Administrator Console, click Protection on the navigation bar.
2. In the Display pane, select the protection group for the SharePoint farm.
3. In the Actions pane, click Modify protection group. This starts the Modify Protection Group
Wizard.
4. On the Select Group Members page, ensure that the node that corresponds to the
SharePoint front-end Web server is marked for selection.
5. Complete the Modify Protection Group Wizard.
6. Run a consistency check for the SharePoint farm.
18
7
Adding or Removing Servers in SharePoint
Farm
DPM uses its protection agents to communicate with the servers that are member of the
SharePoint farm. When you add new servers to the SharePoint farm which contains data that has
to be backed up, ensure that DPM protection agents are installed on those servers.
DPM uses a single front-end Web server to protect the server farm. When you add other front-
end Web servers or remove front-end Web servers other than the server used by DPM, there is
no impact on protection of the farm.
To remove the front-end Web server that DPM is using while continuing protection of the server
farm, see Switching the Front-End Web Server.
No
te
If the front-end Web server that DPM uses to protect the farm is unavailable, use the
following procedure to change the front-end Web server by starting at step 4.
18
8
To
ch
an
ge
the
fro
nt-
en
d
We
b
ser
ver
tha
t
DP
M
us
es
to
pr
ote
ct
the
far
m
1. Stop the SharePoint VSS Writer service on Server1 by running the following command at
a command prompt:
stsadm -o unregisterwsswriter
2. On Server1, open the Registry Editor and navigate to the following key:
HKLM\System\CCS\Services\VSS\VssAccessControl
3. Check all values listed in the VssAccessControl subkey. If any entry has a value data of 0
and another VSS writer is running under the associated account credentials, change the
value data to 1.
4. Install a protection agent on Server2.
C
a
u
t
i
o
n
18
9
You can only switch Web front-end servers if both the servers are on the same
domain.
5. On Server2, at a command prompt, change the directory to DPM installation location\bin\
and run ConfigureSharepoint. For more information about ConfigureSharePoint, see
Using Using ConfigureSharePoint.
6. There is a known issue when the server farm is the only member of the protection group
and the protection group is configured to use tape-based protection. If your server farm is
the only member of the protection group using tape-based protection, to change the front-
end Web server that DPM uses to protect the farm, you must temporarily add another
member to the protection group by performing the following steps:
a. In DPM Administrator Console, click Protection on the navigation bar.
b. Select the protection group that the server farm belongs to, and then click Modify
protection group.
c. In the Modify Group Wizard, add a volume on any server to the protection group. You
can remove this volume from the protection after the procedure is completed.
d. If the protection group is configured for short-term disk-based protection and long-
term tape-based protection, select the manual replica creation option. This avoids
creating a replica for the volume that you are temporarily adding to the protection
group.
e. Complete the wizard.
7. Remove Server1 from the protection group, selecting to retain the replicas on disk and
tape.
8. Select the protection group that the server farm belongs to, and then click Modify
protection group.
9. In the Modify Group Wizard, on the Select Group Members page, expand Server2 and
select the server farm, and then complete the wizard.
A consistency check will start.
10. If you performed step 6, you can now remove the volume from the protection group.
Upgrading SharePoint
If you are moving from an earlier version of SharePoint to a later version of SharePoint 2010 (for
example, upgrading MOSS 2007 to SharePoint 2010), you must reconfigure protection of the
data.
The following are the two scenarios when a SharePoint farm is updated:
• SQL Server hardware upgrade
• When instances of SQL Server do not change
19
0
SQL Server Hardware Upgrade
If the hardware of the computer that is running SQL Server is upgraded, then the databases are
moved from one instance of SQL Server to another. During the upgrade process, both the farms
are online with the older one being read-only (databases are moved in phases). You can protect
the new SharePoint farm during upgrade. To do this, follow these steps:
1. Install DPM protection agents on the new SharePoint farm servers.
2. Protect the SharePoint farm as a new SharePoint farm.
3. Move databases from the old farm to the new farm.
N
o
t
e
DPM will automatically discover the moved databases in the new farm and start
protecting them. DPM generates a warning alert - Farm Configuration Changed for
the old SharePoint farm. You can ignore this alert.
4. When all the databases are moved to the new server, perform Stop protection of the old
SharePoint farm with the retain data option.
19
1
Moving SharePoint Servers Between
Domains
You cannot do the following for protected computers:
• Change the domain of a protected computer and continue protection without disruption.
• Change the domain of a protected computer and associate the existing replicas and recovery
points with the computer when it is re-protected.
We recommend that you do not change the domain of a protected computer. If you must change
the domain of a protected computer, you must complete two tasks:
• Remove the data sources on the computer from protection while the computer retains its
original domain membership.
• Protect the data source on the computer after it becomes a member of another domain.
To
ch
an
ge
the
do
ma
in
me
mb
ers
hip
of
a
pr
ote
cte
d
co
mp
ute
r
19
2
For information about performing tasks involving protection agents and protection groups,
see DPM Help.
To
ren
am
ea
pr
ote
cte
d
co
mp
ute
r
19
3
Improving DPM Recovery Search for
SharePoint Items
The time required for DPM Recoverable Object Search to return recovery points that meet the
specified criteria will increase over time as the number of recovery points grow and the DPMDB
gets more and more fragmented. You can improve the time taken by the search by carrying out
regular maintenance on the DPMDB.
The following table lists the set of tables for which indexes need to be rebuilt for a specific data
source. To improve the performance of the recovery point search for a data source, you need to
rebuild or reorganize the indexes related to that data source.
Rebuilding Indexes
Rebuilding an index drops the index and creates a new one. In doing this, fragmentation is
removed, disk space is reclaimed by compacting the pages using the specified or existing fill
factor setting, and the index rows are reordered in contiguous pages (allocating new pages as
needed). This can improve SQL query performance by reducing the number of page reads
required to obtain the requested data.
Query to rebuild indexes
USE DPMDB
GO
ALTER INDEX ALL ON <tableName> REBUILD
GO
Reorganizing Indexes
Reorganizing an index defragments the leaf level of clustered and nonclustered indexes on tables
and views by physically reordering the leaf-level pages to match the logical order (left to right) of
the leaf nodes. Having the pages in order improves index-scanning performance. The index is
reorganized within the existing pages allocated to it; no new pages are allocated. If an index
spans more than one file, the files are reorganized one at a time. Pages do not migrate between
files.
Reorganizing also compacts the index pages. Any empty pages created by this compaction are
removed providing additional available disk space.
19
4
In some cases, the gain might not be significant. It is also a longer running operation compared to
rebuilding the index.
Query to rebuild indexes
USE DPMDB
GO
ALTER INDEX ALL ON <tableName> REORGANIZE
GO
See Also
ALTER INDEX (Transact-SQL)
19
5
To
dis
abl
ea
pr
ote
cti
on
ag
ent
In This Section
Using Windows Maintenance Tools on SharePoint Servers
Performing SharePoint Protection Management Tasks
Applying Operating System Updates on SharePoint Servers
Running Antivirus Software on SharePoint Servers
See Also
Using Windows Maintenance Tools on the DPM Server
Using Windows Maintenance Tools on File Servers and Workstations
Using Windows Maintenance Tools on Exchange Servers
Using Windows Maintenance Tools on SQL Servers
Using Windows Maintenance Tools on Virtual Server
19
6
Performing SharePoint Maintenance Tasks
If you schedule automatic deletion of inactive Web sites, coordinate the automatic deletion
schedule with the protection schedule to ensure you have a recent copy of the site backed up.
19
7
• DPM Protection Report shows incorrect data for SharePoint farms
• SharePoint Protection not working properly
• SharePoint site or item recovery fails when a shared folder is used as a temporary location
• SharePoint farm protection fails with ID 30111
• SharePoint index backups fail during profile import
• ConfigureSharepoint.exe fails with error code 997
No
te
• To search list items, you must select only Contains as the search string.
• Ensure that you have selected the correct SharePoint farm name.
19
8
On a secondary DPM server (DPM-DR), even
though incremental and express full jobs happen
successfully for a SQL Server database that
belongs to a SharePoint farm, recovery points are
not displayed in the Recovery task area and no
alerts are triggered
In a DPM primary server, if you are protecting a SQL Server database that belongs to a
SharePoint farm, then ensure that you do not protect that database independently on the
secondary DPM server. To resolve this issue, you must identify all such databases, perform Stop
protection with the delete data option, and then re-protect the SharePoint farm.
20
0
In This Section
Performing General Maintenance on Servers Running Virtual Server
Performing Virtual Server Management Tasks
Recovering Virtual Server Data
No
te
If you disable a protection agent for a server that is a cluster node, you should disable the
protection agent for every node of the cluster.
To
dis
abl
ea
pr
ote
cti
on
ag
ent
20
1
In This Section
Using Windows Maintenance Tools on Virtual Server
Applying Operating System Updates on Virtual Server
Running Antivirus Software on Virtual Server
See Also
Using Windows Maintenance Tools on the DPM Server
Using Windows Maintenance Tools on File Servers and Workstations
Using Windows Maintenance Tools on Exchange Servers
Using Windows Maintenance Tools on SQL Servers
Using Windows Maintenance Tools on SharePoint Servers
20
2
software to modify files, making changes that DPM cannot detect. For instructions on configuring
your antivirus software to delete infected files, see the documentation for your antivirus software.
In This Section
Moving Virtual Servers Between Domains
How to Rename Virtual Servers
Renaming Virtual Machines
Moving a Virtual Machine or Virtual Hard Disk
Protecting Application Data on Virtual Machines
20
3
To
ch
an
ge
the
do
ma
in
me
mb
ers
hip
of
a
pr
ote
cte
d
co
mp
ute
r
20
4
We recommend that you do not change the name of a protected computer. If you must change
the name of a protected computer, you must:
• Remove the data sources on the computer from protection (the old computer name).
• Protect the data source on the computer (the new computer name).
To
ren
am
ea
pr
ote
cte
d
co
mp
ute
r
20
5
Moving a Virtual Machine or Virtual Hard Disk
To
mo
ve
a
virt
ual
ma
chi
ne
tha
t is
pr
ote
cte
d
by
DP
M
1. Copy the virtual machine to the new host. For instructions, see "Copying, managing, and
renaming virtual machines" (http://go.microsoft.com/fwlink/?LinkId=95298).
2. Add the copied virtual machine to a protection group.
3. Remove the original virtual machine from the original host. For instructions, see
"Removing virtual machines" (http://go.microsoft.com/fwlink/?LinkId=95299).
4. Stop protection of the original virtual machine.
20
6
data. However, you cannot specifically recover application data from the recovery points for the
virtual machine; you can only recover the entire virtual machine. When you recover the virtual
machine, applications are recovered with all data that was present at the time that the recovery
point was created.
It is not necessary to install a DPM protection agent on a virtual machine to protect it as a virtual
machine on the Virtual Server host.
To recover only application data for applications running in virtual machines, you must install a
protection agent on the virtual machine and select the application data explicitly as a protection
group member.
You can protect both the virtual machines as guests on the Virtual Server host and the application
data within the virtual machines as applications.
For more information about protecting application data, see the topics on protecting specific data
types, such as Exchange Server data or SQL Server data, in DPM 2010 Help
(http://go.microsoft.com/fwlink/?LinkId=196749).
In This Section
How to Recover the Virtual Server Host
How to Recover a Virtual Machine
How to Recover Virtual Machines as Files
20
7
To
rec
ov
er
a
virt
ual
ma
chi
ne
See Also
How to Recover a Virtual Machine
How to Recover Virtual Machines as Files
20
8
How to Recover a Virtual Machine
When you protect a Virtual Server host and its virtual machines, the recoverable items are the
Virtual Server configuration and each virtual machine. You should recover the Virtual Server
configuration before you recover the individual virtual machines.
When you recover the virtual machine, applications are recovered with all data that was present
at the time that the recovery point was created.
To
rec
ov
er
a
virt
ual
ma
chi
ne
20
9
See Also
How to Recover the Virtual Server Host
How to Recover Virtual Machines as Files
To
rec
ov
er
virt
ual
ma
chi
ne
s
as
file
s
21
0
6. On the Review recovery selection page, click Next.
7. Select Copy files to network location, and then click Next.
8. On the Specify destination page, specify the network folder to which the files should be
copied.
9. Specify your recovery options:
a. Select Apply security settings of the destination computer or Apply the security
settings of the recovery point version.
b. Select Enable SAN-based recovery using hardware snapshots to use SAN-based
hardware snapshots for quicker recovery.
This option is valid only when you have a SAN where hardware snapshot
functionality is enabled, the SAN has the capability to create and split a clone to
make it writable, and the protected computer and the DPM server are connected to
the same SAN.
c. In the Notification area, click Send an e-mail when the recovery completes, and
specify the recipients who will receive the notification. Separate the e-mail addresses
with commas.
10. On the Summary page, review the recovery settings, and then click Recover.
See Also
How to Recover the Virtual Server Host
How to Recover a Virtual Machine
Protecting Hyper-V
System Center Data Protection Manager (DPM) 2010 extends DPM support to Microsoft Hyper-V.
DPM now supports both host-based protection for Hyper-V where the agent is installed on the
host computer, and guest-based where the agent is installed directly on the virtual machine.
211
pause a virtual server, take a snapshot of the server, bring the virtual server online again, and
then back up the snapshot.
• Clustering
DPM 2010 supports protection for clustered servers, clustered Hyper-V host server, and
Windows Server Core-based Hyper-V hosts. DPM also supports Virtual Machine Manager
(VMM) Quick Migration.
Unsupported Scenarios
DPM 2010 does not support the backup of virtual machines that do not have storage on the host.
However, if you have at least one VHD of the virtual server on the local machine, DPM protects
the local VHD.
Storage not on host means you either use iSCSI to present volumes to the virtual machine or use
a remote VHD. For such machines, we recommend that customers perform host-level backup of
the VHD files using DPM and install an agent into the virtual machine to back up data that is not
visible on the host.
Ca
uti
on
If you migrate your file server volume using DPM tools and want to continue backups to
the same replica volume, you cannot perform original location recovery for the recovery
points created before the migration. Recovery fails with the message - Couldn't find the
selected volume. You can, however, perform recovery to an alternate location.
See Also
Conditions When DPM Fails to Back up Hyper-V Virtual Machines in an Online State
21
2
• The customer should be running Hyper-V RTM. The corresponding Windows update is
KB950050 (http://go.microsoft.com/fwlink/?LinkID=132735).
• Install the following updates:
• KB951308 (http://go.microsoft.com/fwlink/?LinkID=132733) on each cluster node for
cluster deployment.
Increased functionality and virtual machine control in the Windows Server 2008 Failover
Cluster Management console for the Hyper-V role.
• KB956697 (http://go.microsoft.com/fwlink/?LinkID=133781). This update may be applied
when the Hyper-V writer seems to go missing due to the presence of corrupt Virtual
Machine configuration files in the Hyper-V Server.
• KB958184 (http://go.microsoft.com/fwlink/?LinkId=143332). This update may be applied
when virtual machine files are saved on a volume mounted on a failover cluster using a
volume GUID.
• KB959962 (http://go.microsoft.com/fwlink/?LinkId=136583) on the Hyper-V hosts. This
update for Hyper-V writer is required if the backup fails for one of the following reasons:
• Retryable VSS error.
• VSS application writer provider is going to bad state.
• If recovery of a virtual machine is failing because it has legacy network adapters
attached.
Following this update, the integration services on each of the virtual machines running
on the server must be updated by inserting the Integration Services Disk from the Action
menu in the virtual machine’s management console. This will require the virtual machine
to restart.
• KB960038 (http://go.microsoft.com/fwlink/?LinkId=136584) on all Hyper-V hosts. This
update for Windows Server 2008 fixes a crash of the Hyper-V host server which you may
experience when backups are made using Hyper-V writer.
• The version of Integration Components running inside the VM should be the same as the
version of Hyper-V on the host. For Hyper-V RTM it is 6.0.6001.18016.
You can confirm this in the Device Manager inside the guest VM. Under System Devices in
Device Manager, right-click the entry Hyper-V Volume Shadow Copy and choose Properties.
Check the version under the Driver tab. If the version does not match, insert the integration
services disk by choosing the option under the Action menu in the VM console. Install the
integration components and reboot the VM.
Known Issues
Issue Symptoms Cause Workaround Solution Recommendation
Hyper-V You may experience a This is a N/A Download We recommend
server Stop 0x7E when you problem in and install that you deploy
stops backup a server that the KB960038 this in the next
working has the Hyper-V role volsnap.sys on the available
when installed. driver when Hyper-V downtime window
21
3
DPM runs interacting host to on the Hyper-V
backups with the solve this host.
on the Hyper-V problem.
server. VSS writer.
Backup of Hyper-V virtual . Perform an Download We recommend
Hyper-V machine backups offline and install that you install
virtual (backups using child backup of the KB959962 this update on all
machine partition) fail with virtual on the Hyper-V hosts
fails following error for machine. Hyper-V protected by
multiple volume virtual host to DPM.
machines: DPM solve this
encountered a problem.
retryable VSS error.
(ID 30112 Details:
Unknown error
(0x800423f3)
(0x800423F3)).
On the Hyper-V host
in the event log:
VMMS Event log
contains following
entry:
Failed to revert to
VSS snapshot on one
or more virtual hard
disks of the virtual
machine '%1'. (Virtual
machine ID %2).
Recovery The Hyper-V Virtual This issue Restart the Download We recommend
of Hyper- Machine Management can occur if Hyper-V and install that you deploy
V virtual service stops when the virtual Virtual KB959962 this update on all
machine you are restoring a machine is Machine on the Hyper-V hosts
fails virtual machine. configured Management Hyper-V protected by
to use a service. host to DPM.
legacy solve this
network problem.
adapter. Then
recover
the data to
an
alternate
location.
Backup of Hyper-V VSS writer is This is a To get the Download We recommend
21
4
the in a bad state and problem in writer back to and install that you deploy
Hyper-V does not identify itself the Hyper-V a stable KB959962 this update on all
virtual to VSS anymore. The writer in state, restart on the Hyper-V hosts
machine VSS application writer Windows the Hyper-V Hyper-V protected by
fails or the VSS provider is Server Virtual host to DPM.
in a bad state. Either it 2008. Machine solve this
was already in a bad Management problem.
state or it entered a service.
bad state during the Perform
current operation (ID offline
30111 Details: backups of
Unknown error the virtual
(0x80042319) machines to
(0x80042319)). prevent the
Please check that the writer from
Event Service and the going into a
VSS service are bad state.
running, and check for
errors associated with
these services in the
Application Event Log
on the server
hyper01.contoso.com.
Please allow 10
minutes for VSS to
repair itself and then
retry the operation.
No
te
If the errors persist, review the event log on the target server and on the guest installation
of the VM for which the backups failed for further troubleshooting. Please review errors
with source=VSS.
21
5
Conditions When DPM Fails to Back up
Hyper-V Virtual Machines in an Online State
By default, DPM 2010 performs a backup of a Hyper-V Virtual Machine (VM) in an online state.
However, DPM cannot back up a Hyper-V VM in an online state, if one or more of the following
conditions are true:
• Backup (Volume Snapshot) Integration Service is disabled or not installed.
• VM has one or more dynamic disks.
• VM has one or more non-NTFS based volumes.
• The VM Cluster Resource Group in a cluster setup is offline.
• VM is not in a running state.
• A ShadowStorage assignment of a volume inside the VM is explicitly set to a different volume
other than itself.
These conditions are set by the Hyper-V writer. In such a case, the VM is put in a saved state
before a snapshot of host volumes are taken (except when the VM is turned off) for a backup.
The Hyper-V writer adds the VM in the following format:
• For offline backups: Backup Using Saved State\<VMName>
• For online backups: Backup Using Child Partition Snapshot\<VMName>
No
te
During offline/online backups, the name of the data source remains unchanged even if
the VM configuration changes to support online backups or for any further backups.
21
6
Laptop and notebook computers will not be connected to the network at all times and the number
of protected client computers can be much larger than the number of protected file servers.
These scenarios have resulted in the following changes about how DPM manages client
computer protection.
• The administrator can configure protection for the client computer that they want to protect
without being online. We recommend that administrators use software distribution
mechanisms such as Microsoft System Center Configuration Manager to install and configure
the DPM protection agent.
• The client computer polls the DPM server at 15 minute intervals and obtains the backup
schedule that the administrator specifies for the protection group. The client computer starts
the backup according to the schedule, or by user demand. Alternatively, once the
administrator configures a protection group that allows the end user to specify their protected
data items, the end user can start a backup at any time from the Data Protection Manager
Client.
• DPM will not show alerts for client computers that usually appear for protected servers. These
alerts pertain to failures of individual jobs. For example, a synchronization failure alert will not
appear for the DPM administrator to act upon for any of the failed synchronizations. This is
because client computers are designed to retry the synchronization in the event of a failure.
However, DPM allows you to configure DPM to alert the end user if a client computer has not
been backed up for a predefined number of days that the administrator defined when they
created the protection group.
In This Section
Configuring Client Computer Protection
Protecting Client Computer Data
Recovering Client Computer Data
Performing Client Computer Management Tasks
In This Section
Client Computer Operating System Requirements
Installing Protection Agents on Client Computers
21
7
Client Computer Operating System
Requirements
Before protecting your client computers, you must ensure that the following prerequisites are
installed on the client computer:
N
o
t
e
Recovery from previous versions of files and folders is not supported on computers
running Windows XP.
• Windows Vista or Windows Vista with Service Pack 1 (SP1)
• Windows 7 Client
Network Requirements
Following are the network requirements for DPM 2010 when working with the following types of
client computer connections:
21
8
DPM 2010 Deployment Guide, see Configuring Windows Firewall on the DPM Server
(http://go.microsoft.com/fwlink/?LinkID=179380).
No
te
For backups of client computers that are intermittently connected to the network, and that
are expected to connect over a VPN, we recommend that your Internet connection speed
be a minimum of 1 Megabit per second (Mbps).
DPM supports the following VPN protocols:
• Point-to-Point Tunneling Protocol (PPTP)
• Secure Socket Tunneling Protocol (SSTP)
• Layer 2 Tunneling Protocol (L2TP)
Im
po
rta
nt
To perform backups over a VPN, you must enable Internet Control Message Protocol
(ICMP). For more information, see How to enable ICMP traffic from protected SecureNet
clients to external hosts in ISA Server 2006 and ISA Server 2004
(http://go.microsoft.com/fwlink/?LinkId=186982).
21
9
Installing Protection Agents on Client
Computers
DPM provides a Windows Installer (.msi) file located in the C:\Program Files\Microsoft
DPM\DPM\Agents folder of your DPM server you can use to install a protection agent. You can
deploy the .Windows Installer file using Group Policy or for scalability, we recommend that you
install the protection agents independently using Microsoft System Center Configuration
Manager. To create a Systems Management Server package for the DPM protection agent, you
must provide the following to the Systems Management Server administrator:
• A share to the DpmAgentInstaller.exe and DpmAgentInstaller_AMD64.exe packages.
• A list of computers on which you are installing the protection agents.
• The name of the DPM server.
To
ins
tall
the
pr
ote
cti
on
ag
ent
ma
nu
all
y
on
a
tar
get
co
mp
ute
r
• On the computer on which you want to install the protection agent, we recommend that
you map a network drive to the DPM server.
For example, at the command prompt type net use Z:\\DPM1\c$.
• On the protected computer, from the command prompt change the directory (CD) for the
appropriate processor. For example:
• For 32-bit computers, type CD \<drive letter>:\Program Files\Microsoft
DPM\DPM\Agents\RA\<build number>\i386
For 64-bit computers, type CD \<drive letter>:\Program Files\Microsoft
DPM\DPM\Agents\RA\<build number>\AM64
22
0
• After you change the directory, type DPMAgentInstaller_<processor> <DPM server
name>. For example:
• For 32-bit computers, type DPMAgentInstaller_x86.exe
DPM1.Fully.qualified.domain
• For 64-bit computers, type DPMAgentInstaller_x64.exe <DPM server name>
Note
If you use the DPM server name in the command line, DPM installs the
protection agent and configures the security permissions for the DPM server.
You can perform a manual installation using the /q parameter after the
DpmAgentInstaller.exe command. For example, type:
DpmAgentInstaller.exe /q <DPM server name>.
• If you run the DpmAgentInstaller.exe command from Step 2, to complete the protection
agent configuration for the appropriate DPM server and firewall settings, open an
elevated command prompt, and type:
<drive letter>:\Program Files\Microsoft Data Protection Manager\bin\
SetDpmServer.exe – dpmServerName <DPM server name>.
For example: SetDpmServer.exe –dpmServerName DPM01, where DPM01 is the
actual DPM server name.
If you have already configured protection for the selected clients on the DPM server, DPM will
automatically start the backup for these computers to the DPM server. If you have not yet
22
1
configured protection, you can start the Create New Protection Group Wizard, and then select the
Clients option.
In This Section
Creating a Protection Group on the Client Computer
Adding a Client Computer and Modifying Disk Allocation
22
2
To
ad
da
cli
ent
co
mp
ute
r
usi
ng
the
Cr
eat
e
Ne
w
Pr
ote
cti
on
Gr
ou
p
Wi
zar
d
Comp1.abc.domain.com
Comp2.abc.domain.com
Comp3.abc.domain.com
• If DPM cannot find any of the computers that you specified in the .txt file or that you
entered in the Text file location box, the failed set of computers is placed in a log
file. Click the Failed to add machines link at the bottom of the page to open the log
file.
22
3
4. On the Specify Inclusions and Exclusions page, specify the folders to include or
exclude for protection on the selected computers. To select from a list of well-known
folders, such as Documents, click the drop-down list.
When specifying inclusions and exclusions, note the following:
• When you exclude a folder, and then specify a separate inclusion rule for a subfolder,
DPM does not backup the subfolder. The exclusion rule overrides the inclusion rule.
• When you include a folder, and then specify a separate exclude rule for a subfolder,
DPM backs up the entire folder, except for the excluded subfolder.
• When you include a well-known folder such as Documents, DPM locates the
Documents folder for all users on the computer, and then applies the rule. For
example, if the user profile for computer Comp1 contains the Documents folder for
both User1 and User2, DPM will back up both folders.
a. Type the folder names in the Folder column using variables such as programfiles, or
you can use the exact folder name. Select Include or Exclude for each entry in the
Rule column.
b. Select Allow users to specify protection members to give your end users the
choice to add more folders on the computer that they want to back up. However, the
files and folders you have explicitly excluded as an administrator cannot be selected
by the end user.
c. Under File type exclusions specify the file types to exclude using their file
extensions, and then click Next to continue.
Figure 1 shows an example of how you can use the Specify Inclusions and
Exclusions page to include and exclude specific folders. In this example the My
Documents folder is selected for protection and the Temporary Internet Files folder
is excluded from protection.
<Placeholder for graphic>
5. On the Select Data Protection Method page, in the Protection Group Name box, type
a name for the protection group.
6. In the Protection method section, select if you want to use short-term disk-based
protection or long-term tape-based protection. Click Next to continue.
N
o
t
e
22
4
computers, as well as long-term tape-based protection. DPM does not support
short-term tape-based backup for desktop and laptop computers.
7. On the Specify Goals page, specify your protection goals such as retention range and
synchronization frequency. Select the Alerting option to receive alerts when the
recovery points fails for the selected number of days, and then click Next.
8. On the Allocate Storage page, specify the size of data to be protected on the computer.
We recommend that you co-locate multiple data sources to one DPM replica volume.
N
o
t
e
We recommend that you co-locate your data if you have a large number of client
computers. You will not be able to protect 1000 or more client computers with
one DPM server without co-locating your data. We recommend that you do not
co-locate if you have less than ten client computers in a protection group.
9. Select the Automatically grow the volumes check box to automatically grow volumes
when more disk space is required for protecting data on the client computers. Click Next
to continue.
10. On the Summary page, review your selections and then click Create Group to complete
the wizard.
22
5
To
ad
d
the
cli
ent
co
mp
ute
r
To
mo
dif
y
dis
k
all
oc
ati
on
22
6
To
rec
ov
er
pr
ote
cte
d
dat
a
for
de
skt
op
co
mp
ute
rs
War
ning
22
7
on tape.
8. Click Next after you have specified one of the preceding options.
9. Specify your recovery options:
a. Existing version recovery behavior. Select Create copy, Skip, or Overwrite. This
option is enabled only when you selected Recover to the original location in step 7.
b. Restore security. Select Apply settings of the destination computer or Apply the
security settings of the recovery point version.
c. Network bandwidth usage throttling. Click Modify to enable network bandwidth
usage throttling.
d. Enable SAN based recovery using hardware snapshots. Select this option to use
SAN-based hardware snapshots for quicker recovery.
This option is valid only when you have a SAN where hardware snapshot
functionality is enabled, the SAN has the capability to create a clone and to split a
clone to make it writable, and the protected computer and the DPM server are
connected to the same SAN.
e. Notification. Click Send an e-mail when the recovery completes, and specify the
recipients who will receive the notification. Separate the e-mail addresses with
commas.
10. Click Next after you have made your selections for the preceding options.
11. Review your recovery settings, and click Recover.
N
o
t
e
Any synchronization job for the selected recovery item will be canceled while the
recovery is in progress.
22
8
In This Section
Using the Disk Utilization Report
On some client computers, you may notice the computer running slow when a backup is in
progress. You can improve the computer’s responsiveness using the registry key DWORD
WaitInMSPerRequestForClientRead. This key is not created by default. You must create it at
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft Data Protection
Manager\Agent\ClientProtection.
The default value for this DWORD is 50 (32H). You can increase it to 75 or 100 to improve
responsiveness. If you want to increase backup speed at the expense of responsiveness, reduce
the value to 40 or 30.
In This Section
Prerequisites for Hyper-V Protection
22
9
Protecting Hyper-V Machines
Recovering Hyper-V Data
Understanding Protection for CSV
No
te
DPM continues to protect Hyper-V virtual machines even if the Hyper-V role is not
installed on the DPM server. However, you cannot do item-level recovery (ILR) unless the
Hyper-V role is enabled.
No
te
Things to Remember
• You must turn on auto-mount on the host computer to enable virtual machine protection.
• You must disable TCP Chimney Offload.
Caching
When you are protecting a large number (over 200 virtual machines) of virtual machines using
cluster shared volumes (CSV), it may take a long time (>15 min) to populate the inquiry screen in
the Create New Protection Group Wizard. You can avoid this by enabling caching on the primary
DPM server. After you enable caching, when you expand the data source on the inquiry screen,
23
1
DPM will refresh the resource groups under that node, but the virtual machines under each
resource group will be populated from the cache. The default time-out for the cache is 48 hours.
You can change this by editing the registry.
To enable caching, create a new registry key called CacheInquiryResults at
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft Data Protection
Manager\Configuration. Under this key, create the DWORD InquiryResultsTimeoutInterval.
After you create the registry key, you will see the Clear Cache button on the inquiry page of the
Create New Protection Group Wizard. Click this button when you want to force DPM to refresh
the list of virtual machines.
Wa
rni
ng
When you click Clear Cache, DPM will refresh the entire cache, and not just the cache of
the selected resource group.
23
2
Server Name Fully Qualified Domain Name hyperv01.contoso.com
of the Hyper-V host server.
Protection Group Name of the existing Protection Group 3
protection group to which we
are adding the new virtual
machines.
Scenario Description
Recovering a virtual machine to an alternate The original VHD is deleted. DPM will recover
location the VHD and other configuration files on the
original location by using the Hyper-V VSS
writer. At the end of the recovery process,
virtual machines will still be highly available.
23
3
Wa
rni
ng
Tip
In This Section
Recovering A Hyper-V Machine To Its Original Location
Recovering a Virtual Machine to an Alternate Location
Item-Level Recovery for Hyper-V
23
4
Recovering A Hyper-V Machine To Its
Original Location
The procedure to recover a Hyper-V virtual machine to its original location is the same as with
any other data source. For more information, in DPM 2010 Help, see Recovery Wizard
(http://go.microsoft.com/fwlink/?LinkID=164845).
Wa
rni
ng
When you recover to an alternate clustered host, DPM will not make the virtual machine
highly available. You must do that using the Failover Cluster Manager.
System Center Data Protection Manager (DPM) 2010 also supports alternate location recovery of
Hyper-V virtual machines to a cluster both in a Cluster Share Volume (CSV) and Non-CSV
environment.
In This Section
Procedure To Recover A Hyper-V Virtual Machine In A Non-CSV Environment
Procedure To Recover To An Alternate Stand-Alone Hyper-V Host
23
5
Procedure To Recover A Hyper-V Virtual
Machine In A Non-CSV Environment
W
a
r
n
i
n
g
If you select the virtual machine on the left pane, the Recoverable Item list will show
you the list of VHDs. If you do a recovery at this point, you are recovering a VHD and
not the virtual machine.
5. In the Actions pane, click Recover. DPM starts the Recovery Wizard.
For more information about the Recovery wizard, see Recovery Wizard.
23
6
W
a
r
n
i
n
g
23
7
W
a
r
n
i
n
g
If you select the virtual machine on the left pane, the Recoverable Item list will show
you the list of VHDs. If you do a recovery at this point, you are recovering a VHD and
not the virtual machine.
6. In the Actions pane, click Recover. DPM starts the Recovery Wizard.
For more information about the Recovery wizard, see Recovery Wizard.
7. You can now close the Recovery Wizard and view the recovery status in the Monitoring task
area.
W
a
r
n
i
n
g
• If the recovered virtual machine was backed up in an online state, and it is saved state
after recovery, delete the saved state of that virtual machine from the Hyper-V Manager
Console and start it.
• After alternate location recovery always check whether the virtual machine’s network
configuration is correct.
23
8
DPM supports item-level recovery (ILR), which allows you to do granular recovery of files, folders,
volumes, and virtual hard disks (VHDs) from a host-level backup of Hyper-V virtual machines to a
network share or a volume on a DPM protected server.
You must have the Hyper-V role enabled on the DPM server to perform item-level recoveries.
During item-level recovery, DPM has to mount the VHDs of the protected virtual machines.
Im
po
rta
nt
Item-level recovery does not support recovery of an item to its original location.
You cannot do an item-level recovery from within a virtual machine.
The following table lists supported and unsupported scenarios when recovering files, folders,
volumes, and VHDs using ILR in a Hyper-V virtual machine.
Im
po
rta
nt
Mount points cannot be traversed or browsed when exploring a VHD for item-level
recovery.
23
9
In This Section
Procedure for Item-Level Recovery of Files and Folders
Procedure for Item-Level Recovery of Volumes
Procedure for Item-Level Recovery of VHD
To
per
for
m
ite
m-
lev
el
rec
ov
ery
of
file
s
an
d
fol
der
s
N
o
t
e
24
0
Dates in bold indicate available recovery points.
4. To view the list of files and folders, in the Recoverable Items list, do the following:
a. Double-click the item (VHD) that you want to recover.
b. Double-click the items (volumes in VHD) that you want to recover.
5. Select the items (files and folders) that you want to recover.
N
o
t
e
You can select and recover multiple files and folders from the list.
6. In the Actions pane, click Recover to start the Recovery Wizard.
For more information about the Recovery Wizard, see Recovery Wizard
(http://go.microsoft.com/fwlink/?LinkID=164845).
N
o
t
e
DPM saves files and folders in a custom directory structure in the following
format: <Recovery destination selected by user>\<VM name>_<Backup Time
stamp> with the exact file system hierarchy that is used on a protected computer
with the DPM agent installed.
24
1
Procedure for Item-Level Recovery of
Volumes
Use this procedure for item-level recovery of volumes.
To
per
for
m
ite
m-
lev
el
rec
ov
ery
of
vol
um
es
N
o
t
e
The list pane displays the volume label or “Virtual Machine Volume” if no volume
label is available.
You cannot select and recover multiple volumes at the same time.
5. In the Actions pane, click Recover to start the Recovery Wizard.
24
2
For more information about the Recovery Wizard, see Recovery Wizard
(http://go.microsoft.com/fwlink/?LinkID=164845).
To
per
for
m
ite
m-
lev
el
rec
ov
ery
of
virt
ual
har
d
dri
ve
s
(V
HD
)
N
o
t
e
24
3
Recoverable Items pane.
4. In the Recoverable Items pane, select the item (VHD file) that you want to recover.
N
o
t
e
You cannot select and recover multiple VHDs at the same time.
5. In the Actions pane, click Recover to start the Recovery Wizard.
N
o
t
e
When you are recovering a VHD of a virtual machine that has Hyper-V
snapshots, .AVHD files are not displayed in the Recoverable Items pane, but
DPM will recover the parent VHD and all the associated .AVHD files.
DPM saves VHDs in a custom directory structure in the following format:
DPM_<backup-time>\DPM_Recovered_At_<RecoveryTime>\<Path of the VHD
on the protected computerr> with the exact file system hierarchy that is used on a
protected computer with the DPM agent installed.
For more information about the Recovery Wizard, see Recovery Wizard
(http://go.microsoft.com/fwlink/?LinkID=164845).
24
4
In this example, when VM2 is being backed up, the CSV is made local to Node B. All I/O for VM1
is then routed over the network through the CSV filter on Node B. This affects the performance of
VM1 and it appears on the Failover Cluster Manager as being in “redirected I/O mode.”
To reduce the impact on VM1, we recommend that you use hardware snapshots, which enable
the CSV to resume direct I/O mode as soon as the hardware snapshot has been created. The
duration of this process is typically very short, about two minutes.
24
5
If you use software snapshots, the CSV will be in redirected I/O mode for all virtual machines on
this CSV, other than VM2, for the duration of the backup. The duration of this process depends on
the size of the VHDs being backed up by DPM, and can be significant. Therefore, using software
snapshots can decrease the performance of the virtual machines.
After the snapshot is created, DPM starts replicating the data from the snapshot to the DPM
server. After the replication is complete, the protection agent deletes the hardware snapshot.
In This Section
Considerations for Backing Up Virtual Machines on CSV with Hardware VSS Providers
Migrating from the System VSS Provider to a Hardware VSS Provider
Considerations for Backing Up Virtual Machines on CSV with the System VSS Provider
24
6
Key HKLM\Software\Microsoft\Microsoft Data Protection Manager\2.0\Configuration\MaxAllowe
dParallelBackups
Valu Microsoft Hyper-V
e
Data 3
Type DWORD
This will enable a maximum of three backups to run concurrently on each node. For optimal
performance, we recommend that you do not use a value greater than 3.
4. To complete this process, you must run the Modify Protection Group Wizard for each of the
protection groups that protect the virtual machines on this cluster.
24
7
Considerations for Backing Up Virtual
Machines on CSV with the System VSS
Provider
If your SAN vendor does not have hardware VSS providers, you can use software snapshots to
back up your virtual machines.
We recommend that virtual machines deployed on CSV should be backed up serially.
There are two aspects to serialization of backup jobs in a CSV environment:
• Serializing virtual machine backups on a per node basis.
• Serializing virtual machine backups on a per CSV LUN basis.
This ensures that only one backup job will run at a time on a Hyper-V host.
$infoText = "This script will generate the DatasourceGroups.xml file in the current path.
Once this file is created merge it with the same file name under %programfiles%\Microsoft
DPM\DPM\Config directory on the DPM server. Read the documentation for more details."
24
8
echo $infoText
$footer = "</DatasourceGroup>"
$dir = [guid]::NewGuid()
md $dir
$cluster = get-Cluster
echo $line >> $dir\$vol # File VolumeX will contain entries for all VMs hosted on CSV
VolumeX
$GroupEndString = "</Group>"
24
9
{
25
0
3. This script will generate the DataSourceGroups.xml file in the current path. After this file is
created, copy it to the %programfiles%\Microsoft DPM\DPM\Config directory on the DPM
server.
4. You can verify the groupings by opening the XML file that is generated. The following is the
expected format:
<?xml version="1.0" encoding="utf-16"?>
<DatasourceGroup xmlns:xsi="http://www.w3.org/2001/XMLSchema-
instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"
xmlns="http://schemas.microsoft.com/2003/dls/GroupDatasourceByDi
sk.xsd">
<Group GroupName="Group1">
<Datasource DatasourceName="EA24071A-7B7B-42CF-AB1D-
BBAE49F50632" ProtectedServerName="SCVMM VM-Vol7-03
Resources.CSVSCALE.SCALEDPM01.LAB" WriterId="66841cd4-6ded-4f4b-
8f17-fd23f8ddc3de" />
</Group>
</DatasourceGroup>
No
te
You can skip this step if the DPM server is protecting only one CSV cluster. The
generated DataSourceGroups.xml file can be used directly on the DPM server.
1. Copy any one of the DataSourceGroups.xml files that was generated to the DPM server
under the location %Programfiles%\Microsoft DPM\DPM\Config.
2. Open the file to edit it.
<?xml version="1.0" encoding="utf-16"?>
<DatasourceGroup xmlns:xsi="http://www.w3.org/2001/XMLSchema-
instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"
xmlns="http://schemas.microsoft.com/2003/dls/GroupDatasourceByDi
sk.xsd">
<Group GroupName="Group1">
25
1
<Datasource DatasourceName="EA24071A-7B7B-42CF-AB1D-
BBAE49F50632" ProtectedServerName="SCVMM VM-Vol7-03
Resources.CSVSCALE.SCALEDPM01.LAB" WriterId="66841cd4-6ded-4f4b-
8f17-fd23f8ddc3de" />
</Group>
</DatasourceGroup>
3. Copy the <Group> tags from all the DataSourceGroup.xml files generated and add the text
between the <DataSourceGroup> tags. The DataSourceGroups.xml file will now contain one
<header> tag, one <DataSourceGroup> tag, and <Group> tags from all CSV clusters.
4. Close the DataSourceGroups.xml file on the DPM server. It is now ready to use.
Supported Scenarios
Workgroup Untrusted Domain
Files – Basic - All server and client Supported Supported
SKUs
Files – Clustering Not applicable Not Supported
System State – Windows Supported Supported
Server 2003, Windows
Server 2008, Windows 2008 R2
SQL Server– Basic – SQL Supported Supported
Server 2000, SQL Server 2005,
SQL Server 2008
25
2
SQL Server - Mirroring Not Supported Not Supported
SQL Server - Clustering Not applicable Not Supported
Hyper-V – Basic – Windows Supported Supported
Server 2008, Windows 2008 R2
Hyper-V – Clustering Not applicable Not Supported
Hyper-V – Cluster Shared Volume Not applicable Not Supported
Exchange – Basic – Exchange Not applicable Supported
Server 2003, Exchange
Server 2007, Exchange
Server 2010
Exchange Server – Clustering Not applicable Not Supported
Exchange Server – CCR Not applicable Not Supported
Exchange Server – LCR Not applicable Supported
Exchange Server – SCR Not applicable Not Supported
Exchange Server – DAG Not applicable Supported
Microsoft SharePoint Server Not Supported Not Supported
Portable computers Not Supported Not Supported
Bare Metal Recovery Not Supported Not Supported
End User Recovery Not Supported Not Supported
DPM Disaster Protection of Not Supported Not Supported
workgroup/untrusted domain
computers
25
3
• Default Port: 5718 for agent coordinator;
5719 for protection agent.
• Authentication: NTLM, using credentials
specified after DPM agent installation
DPM account requirements Local account without administrative rights on
the production server.
Use NTLM v2 for secure communication
between DPM and protected computer.
Agent installation Requires local installation of the DPM agent on
the protected computer and running
SetDpmServer. After installing the agent, use
the Install Agent Wizard to attach the
production server to DPM.
Restrictions • SharePoint and disconnected client
protection is not supported in DPM 2010.
• DPM disaster recovery is not supported in
DPM 2010.
• Clustering/mirroring for Files/SQL
Server/Exchange Server is not supported in
DPM 2010.
• Protection of perimeter network (DMZ)
machines is not supported in DPM 2010.
Im
po
rta
nt
Make sure IPSEC does not block communication between DPM server and workgroup
machines.
25
4
Im
po
rta
nt
To protect a computer that is running Windows XP, you must first disable the ForceGuest
registry key otherwise NTLM authentication will fail while attaching the computer.
For more information about disabling the ForceGuest registry key, see How to Set
Security in Windows XP Professional That Is Installed in a Workgroup
(http://go.microsoft.com/fwlink/?LinkID=192212).
Parameter Description
-IsNonDomainServer Specifies that this server is in a workgroup
or an untrusted domain.
-UserName Creates an NT user account with the specified
username for this server to communicate with
DPM server. This option should be used along
with -IsNonDomainServer.
-ProductionServerDnsSuffix In case there are multiple DNS suffixes
configured for this server,
ProductionServerDnsSuffix represents the DNS
suffix which DPM server will use to
communicate with this server.
-DpmServerName Name of the DPM server. FQDN if DPM server
and protected computer are accessible to each
other using FQDNs. NETBIOS if DPM server
and protected computer are accessible to each
other using NETBIOS names.
25
5
Attaching a Workgroup Computer to the DPM
Server
The steps to attach a workgroup computer using DPM Administrator Console are as follows.
1. Start the Protection Agent Installation Wizard from the DPM Administrator Console.
2. Select Attach and click Next.
3. Enter the computer name, user name, and password for the computer that you want to attach
to. This should be the same as the login credentials specified during agent installation on that
computer. Click Next.
4. Review the information on the Summary page, and then, if the information is correct, click
Install. After the attach action is completed successfully, click Close.
This script registers the specified workgroup computer to be protected with this DPM server,
creates a local user account using the specified credentials, and configures DPM to use these
credentials to authenticate the workgroup computer.
Im
po
rta
nt
Before attaching the workgroup computer to the DPM server by using the DPM
Administrator Console or DPM Management Shell, you must install the DPM agent and
run SetDpmServer.exe on the workgroup computer.
25
6
Im
po
rta
nt
If you use the NetBIOS name of the DPM server in the SetDPMServer command, you
also must use the NetBIOS for the protected computer when you attach the computer.
This also applies if you use the fully qualified domain name (FQDN) of the DPM server.
Examples
Example 1
Configuring a workgroup computer for protection after agent is installed.
On the workgroup computer, run SetDpmServer.exe -DpmServerName Server01 -isNonDomainServer
-UserName mark.
Im
po
rta
nt
Because the workgroup computers are typically accessible only by using NetBIOS name,
the value for DPMServerName must be the NetBIOS name.
Example 2
Configuring a workgroup computer with conflicting NetBIOS names for protection after agent is
installed.
On the workgroup computer, run SetDpmServer.exe -dpmServerName Server01.corp.contoso.com
-isNonDomainServer -userName mark -productionServerDnsSuffix widgets.corp.com.
25
7
Protecting Computers on Untrusted Domains
Parameter Description
-IsNonDomainServer Specifies that this server is in a workgroup
or an untrusted domain.
-UserName Creates an NT user account with the specified
username for this server to communicate with
DPM server. This option should be used along
with -IsNonDomainServer.
-ProductionServerDnsSuffix In case there are multiple DNS suffixes
configured for this server,
ProductionServerDnsSuffix represents the DNS
suffix which DPM server will use to
communicate with this server.
-DpmServerName Name of the DPM server. FQDN if DPM server
and protected computer are accessible to each
other using FQDNs. NETBIOS if DPM server
and protected computer are accessible to each
other using NETBIOS names.
25
8
3. Enter the computer name, user name and password for the computer you want to attach to.
This should be the same as the login credentials specified during agent installation on that
computer. Click Next.
4. Review the information on the Summary page and click Install if the information is correct.
Click Close once the attach action is successful.
This script registers the specified computer to be protected with this DPM computer, creates a
local user account using the specified credentials and configures DPM to use these credentials to
authenticate to the computer.
Im
po
rta
nt
DPM agent must be installed and SetDpmServer.exe must be run on the computer,
before attaching the computer to DPM server using the DPM Administrator Console or
Management shell.
Im
po
rta
nt
If you use NetBIOS name of the DPM server in the SetDPMServer command, you must
use the NetBIOS for the protected computer also during attach and vice versa if you are
using FQDN.
25
9
Updating Password for Workgroup or
Untrusted Computers
When you install an agent locally on a workgroup computer, you specify the credentials to
SetDpmServer to generate a local account and DPM uses these credentials to communicate with
the agent on the workgroup computer.
I
m
p
o
r
t
a
n
t
You must use the same naming convention (FQDN or NetBIOS) as you did when
configuring protection.
2. On the DPM server, run the Update-NonDomainServerInfo cmdlet and provide appropriate
information along with new password.
3. Refresh the agent information for the protected computer.
Examples
Example 1
Changing the password when the computer was protected using NetBIOS name.
On the protected computer, run SetDpmServer.exe -dpmServerName Server01 -isNonDomainServer
–UpdatePassword
When prompted, provide the same password as the one you provided in Step 1.
26
0
Example 2
Changing the password when the computer was protected using FQDN.
On the protected computer, run SetDpmServer.exe -dpmServerName Server01.corp.contoso.com
-isNonDomainServer -UpdatePassword
When prompted, provide the same password as the one you provided in Step 1.
Prerequisites
• Windows Server Backup installed on the protected computer for BMR.
• Windows Recovery Environment (WinRE) for BMR.
Unsupported Scenarios
The following scenarios are not supported for BMR:
• Computers running Windows Server 2003.
• Computers running client operating systems like Windows XP or Windows Vista or
Windows 7.
• A DPM server cannot protect itself for BMR.
• Disk-Tape protection is not supported for BMR. However, long-term to tape with short-term to
disk (D-D-T) is supported.
26
1
In This Section
Prescriptive Guidance on BMR vs System State By Data Source
Setting Up BMR Protection
Setting Up System State Protection
Recovering BMR
Recovering System State
Migrating Between System State and BMR Protection
File Servers
• File system backup for data protection.
• BMR backup for system protection.
Recovery Strategy
Lost file data File recovery using DPM
Lost or damaged operating system System State recovery using BMR backup
Lost server (data volumes intact) BMR recovery using BMR backup
Lost server (data volumes also lost) BMR recovery followed by file recovery
SharePoint Farm
• SharePoint farm backup for farm data.
• BMR backup on Web front-end server to protect IIS role.
• BMR or System State backup for servers hosting content database.
Recovery Strategy
Lost site, lists, list items, documents SharePoint recovery using DPM
Lost or damaged operating system System State recovery using BMR backup
26
2
Disaster recovery BMR recovery using BMR backup
Recovery Strategy
Lost virtual machine Use Hyper-V recovery features
Lost or damaged operating system System State recovery using BMR backup
Lost Hyper-V host (virtual machine intact) BMR recovery using BMR backup
Lost Hyper-V host (virtual machine also lost) BMR recovery followed by Hyper-V recovery
Recovery Strategy
Lost application data Application-specific recovery using DPM
Lost or damaged operating system System State recovery using BMR backup
Lost server (database and transaction log files BMR recovery using BMR backup
intact)
Lost server (database and transaction log files BMR recovery followed by application-specific
also lost) recovery
Tip
If your application is installed on a critical volume, you will also be able to restore the
application as part of BMR. However, application data is not backed up as part of BMR.
You can set up BMR protection for a computer by using the Create New Protection Group Wizard.
You can select BMR protection from under the System Protection node on the Select Group
Members page of the wizard.
26
3
Space Requirements
Unlike System State protection, DPM does not have any space requirements on the protected
computer for BMR protection. WSB directly transfers the backups to the DPM server.
Wa
rni
ng
DPM will not show you the progress of this job in the Jobs view.
DPM reserves 30 GB of space on the replica volume for BMR. You can change this by using the
Disk Allocation page in the Modify Protection Group Wizard or the Get-
DatasourceDiskAllocation and Set-DatasourceDiskAllocation cmdlets.
On the recovery point volume, BMR protection requires about 6 GB for retention of five days.
No
te
DPM does not calculate the size of BMR data source, but assumes 30 GB for all servers.
Admins should change the value as per the size of BMR backups expected on their
environments.
Size of BMR backup can be roughly calculated sum of used space on all critical volumes.
Critical volumes = Boot Volume + System Volume + Volume hosting system state data
such as AD DIT/log volumes.
26
4
Wa
rni
ng
When you stop protection for BMR, System State protection is not stopped automatically.
You must specifically clear the System State check box to stop System State protection.
Things to Remember
• You cannot protect BMR and System State for the same computer on different protection
groups.
• You cannot reduce the replica volume size to less than 15 GB.
See Also
Managing System Protection
Tip
We recommend that you protect BMR for complete protection of your computer.
Space Requirements
For System State protection, WSB first creates a local dump of the System State information and
then transfers it to the DPM server. The local dump will typically require 15 GB of space on the
computer. If there is insufficient space on the computer, WSB will fail the backup.
26
5
Wa
rni
ng
DPM will show the progress of this job in the Jobs view only when data transfer begins.
Things to Remember
You cannot protect System State and BMR for the same computer on different protection groups.
See Also
Managing System Protection
Recovering BMR
26
6
W
a
r
n
i
n
g
Computers in WinPE cannot connect to network shares that have IPsec enabled. The
computer should be an IPsec boundary computer so that a computer that is not
joined to the domain can access the network share by using a username and
password.
2. Start the protected computer using Windows Recovery Environment (WinRE) and go to the
command shell.
3. Using the command shell, enable networking - start /w wpeinit
4. Using the command shell, retrieve the version of the backup from the restored image -
Wbadmin get versions –backuptarget:\\<computername>\serverbackup$
5. Using the command shell, start system recovery - Wbadmin.exe start sysrecovery –
version:<version ID from Step 2> -backuptarget:\\<computername>\ServerBackup$
-recreatedisks
See Also
Managing System Protection
Recover the Operating System or Full Server
Windows Server Backup 2008 Restore from Network Location
26
7
W
a
r
n
i
n
g
See Also
Managing System Protection
Recover the System State
26
8
BMR protection will require more space on the replica volume. The volume will be extended
automatically. If you want to change the default space allocations you can use Modify-
DiskAllocation.
Wa
rni
ng
Disaster Replica will fail because of increased space needs. You must manually increase
the space allocation on the server.
Im
po
rta
nt
Because of the increased space requirement on the replica volume, DPM may try to
automatically grow the volume. If there is insufficient space in the storage pool, you will
see an error indicating this.
Tip
If you are trying to remove BMR protection to free up disk space, you must stop
protection of BMR and System State.
See Also
Managing System Protection
26
9
Setting Up Disaster Recovery
System Center Data Protection Manager 2010 (DPM) allows you to protect your data sources on
a secondary DPM server, preferably at a remote location, as a backup to your primary DPM
server. A disaster can take the following forms:
• The primary DPM server and the protected computers are lost.
• Only the primary DPM server is lost.
In the first case, having a secondary DPM server in a remote location allows you to recover your
protected computers quickly. In the second case, you can switch protection so that the secondary
DPM server takes over at the primary DPM server for the protected computers until another
computer can be installed as the primary DPM server.
N
o
t
e
On the secondary DPM server, the wizard does not differentiate between BMR and
System State. If either is protected it will appear as System Protection.
Im
po
rta
nt
We strongly recommend that you protect the DPMDB on the secondary DPM server.
27
0
Wa
rni
ng
If your primary DPM server is protecting a SharePoint farm, you must make sure that you
give the secondary server enough time to back up the primary server before the next
back up is triggered. Using the Modify Protection Group Wizard, you can define when you
want your back up to start,
See Also
Specify Short-Term Goals
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize\Tcp1323Opts
For example, using the following settings over a 100 Mbps link with 40 ms latency produces the
following results:
Settings
On the remote DPM server: 524288
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize
On both DPM servers: 3
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize\Tcp132
3Opts
27
1
Results
One job running 3.45 MB/sec
Three jobs running ~3.00 MB/se
c per job
Using DPMSync
DpmSync is a command-line tool that enables you to synchronize the DPM database with the
state of the disks in the storage pool and with the installed protection agents. The DpmSync tool
restores the DPM database, synchronizes the DPM database with the replicas in the storage
pool, restores the Report database, and reallocates missing replicas.
DpmSync Syntax
DpmSync –RestoreDb –DbLoc location –InstanceName server\instance]
DpmSync -Sync
DpmSync -ReallocateReplica
DpmSync -DataCopied
Parameters
Parameter Description
-RestoreDb Restores a DPM database from a specified
location.
-Sync Synchronizes restored databases.
You must run DpmSync –Sync after you restore
the databases.
After you run DpmSync –Sync, some replicas
may still be marked as missing.
-DbLoc location Identifies the location of backup of DPM
database.
-InstanceName server\instance Instance to which DPMDB must be restored.
-ReallocateReplica Reallocates all missing replica volumes without
synchronization.
-DataCopied Indicates that you have completed loading data
into the newly allocated replica volumes.
This is applicable for client computers only.
Example 1: To restore the DPM database from local backup media on the DPM server.
Run the following command:
DpmSync –RestoreDb -DbLoc G:\DPM\Backups\2005\November\DPMDB.bak
27
2
After you restore the DPM database, to synchronize the databases, you run the following
command:
DpmSync -Sync
After you restore and synchronize the DPM database and before you restore the replica, you run
the following command to reallocate disk space for the replica:
DpmSync -ReallocateReplica
27
3
instancename is the name of the remote SQL
Server instance.
6. Now run the following command on
DPMServer1 Dpmsync –sync
Managing Performance
The topics in this section define performance expectations and explain how to optimize Data
Protection Manager (DPM) performance. Network speed, the performance characteristics of the
protected computer, the size of your protected data, and the rate at which the protected data
changes will determine your actual results.
In This Section
How DPM Operations Affect Performance
DPM and Memory
Performance Counters
Improving Performance
Managing DPM Performance on a WAN
How Protection Group Changes Affect Jobs
See Also
Disaster Recovery
Managing DPM Servers
Managing Protected File Servers and Workstations
Managing Protected Servers Running Exchange
Managing Protected Servers Running SQL Server
Managing Protected Servers Running SharePoint
Managing Protected Virtual Servers
Managing Tapes
In This Section
Replica Creation
Change Tracking
Synchronization
Consistency Check
Express Full Backup
Backup to Tape
DPM Processes
See Also
Managing Performance
Replica Creation
In DPM, a replica is a complete copy of the protected data on a single volume, database, or
storage group. The DPM protection agent on the protected computer sends the data selected for
protection to the DPM server. A replica of each member in the protection group is created.
Replica creation is one of the more resource-intensive DPM operations, with its greatest impact
being on network resources.
Typically, the performance of the replica creation will be limited by the speed of the network
connection between the DPM server and the protected computers. That is, the amount of time
that it takes to transfer a 1-gigabyte (GB) volume from a protected computer to the DPM server
will be determined by the amount of data per second that the network can transmit.
The following table shows the amount of time it would take, at different network speeds, to
transmit various amounts of data under optimal conditions. Times are given in hours, except
where specified as minutes.
27
5
1 GB < 1 minute < 1 hour <1 <1 1.5 6
50 GB <10 minutes 1.5 hour 5 18 71 284
200 GB <36 minutes 6 hours 18 71 284 1137
500 GB <1.5 hours 15 45 178 711 2844
No
te
In the preceding table, Gbps = gigabits per second, Mbps = megabits per second, and
Kbps = kilobits per second. The figures for a network speed of 1 Gbps assume that the
disk speed on the DPM server and the protected computer are not a bottleneck. Typically,
the time to complete initial replica (IR) creation can be calculated as follows:
IR: hours = ((data size in MB) / (.8 x network speed in MB/s)) / 3600
Note 1: Convert network speed from bits to bytes by dividing by 8.
Note 2: The network speed is multiplied by .8 because the maximum network efficiency is
approximately 80%.
On an extremely fast network, such as a gigabit connection, the speed of replica creation will be
determined by the disk speed of the DPM server or that of the protected computer, whichever is
slower.
The impact of replica creation on network performance can be reduced by using network
bandwidth usage throttling. For more information, see Using Network Bandwidth Usage
Throttling.
To avoid the network load of replica creation, you can create replicas manually from tape or other
removable media when creating the initial replica, which can take from hours to days depending
on the amount of data to protect. For more information, see Creating Replicas Manually.
If the network goes down during synchronization, DPM will attempt to continue the
synchronization from the point where it left off last. If the network goes down during consistency
check, DPM will attempt to continue the check if the network comes back up in five minutes.
However, if the network remains down for longer than 5 minutes the replica is marked as
Inconsistent.
See Also
How DPM Operations Affect Performance
Managing Performance
27
6
Change Tracking
After the replica is created, the DPM protection agent on the computer begins tracking all
changes to protected data on that computer. Changes to files are passed through a filter before
being written to the volume. This process is similar to the filtering of files through antivirus
software, but the performance load of DPM tracking changes is less than the performance load of
antivirus software.
See Also
How DPM Operations Affect Performance
Managing Performance
Synchronization
Synchronization is the process by which DPM transfers data changes from the protected
computer to the DPM server and then applies the changes to the replica of the protected data.
For a file volume or share, the protection agent on the protected computer tracks changes to
blocks, using the volume filter and the change journal that is part of the operating system to
determine whether any protected files were modified. DPM also uses the volume filter and
change journal to track the creation of new files and the deletion or renaming of protected files.
For application data, after the replica is created, changes to volume blocks belonging to
application files are tracked by the volume filter.
How changes are transferred to the DPM server depends on the application and the type of
synchronization. For protected Microsoft Exchange data, synchronization transfers an
incremental Volume Shadow Copy Service (VSS) snapshot. For protected Microsoft SQL Server
data, synchronization transfers a transaction log backup.
DPM relies on synchronization to update replicas with the protected data. Each synchronization
job consumes network resources and can therefore affect network performance.
The impact of synchronization on network performance can be reduced by using network
bandwidth usage throttling and compression. For more information, see Using Network
Bandwidth Usage Throttling and Using On-the-Wire Compression.
See Also
How DPM Operations Affect Performance
Managing Performance
27
7
Consistency Check
A consistency check is the process by which DPM checks for and corrects inconsistencies
between a protected data source and its replica.
The performance of the protected computer, DPM server, and network will be affected while a
consistency check is running, but it is expected to be optimized because only the changes and
checksums are transferred.
The network impact from a consistency check is significantly lower than initial replica creation
after a successful replica creation. If the initial replica creation is interrupted or unsuccessful, the
first consistency check can have an impact similar to replica creation.
We recommend that consistency checks be performed during off-peak hours.
DPM automatically performs a consistency check in the following instances:
• When you modify a protection group by changing the exclusion list.
• When a daily consistency check is scheduled and the replica is inconsistent.
See Also
How DPM Operations Affect Performance
Managing Performance
See Also
How DPM Operations Affect Performance
Managing Performance
Backup to Tape
When DPM backs up data from the replica to tape, there is no network traffic and therefore no
performance impact on the protected computer.
27
8
When DPM backs up data from the protected computer directly to tape, there will be an impact on
the disk resources and performance on the protected computer. The impact on performance is
less when backing up file data than when backing up application data.
See Also
How DPM Operations Affect Performance
Managing Performance
DPM Processes
On the DPM server, three processes can impact performance:
• DPM protection agent (MsDpmProtectionAgent.exe). DPM jobs affect both memory and
CPU usage by the DPM protection agent. It is normal for CPU usage by
MsDpmProtectionAgent.exe to increase during consistency checks.
• DPM service (MsDpm.exe). The DPM service affects both memory and CPU usage.
• DPM Administrator Console (an instance of Mmc.exe). DPM Administrator Console can
be a significant factor in high memory usage. You can close it when it is not in use.
No
te
Memory usage for the DPM instance of the SQL Server service
(Microsoft$DPM$Acct.exe) is expected to be comparatively high. This does not indicate a
problem. The service normally uses a large amount of memory for caching, but it
releases memory when available memory is low.
See Also
How DPM Operations Affect Performance
Managing Performance
27
9
DPM requires a pagefile size that is 0.2 percent the size of all recovery point volumes combined,
in addition to the recommended size (generally, 1.5 times the amount of RAM on the computer).
For example, if the recovery point volumes on a DPM server total 3 TB, you should increase the
pagefile size by 6 GB.
For more information about modifying the pagefile size, see Change the size of the virtual
memory paging file (http://go.microsoft.com/fwlink/?LinkId=95116).
There is a Volume Shadow Copy Service (VSS) non-paged pool limitation on 32-bit operating
systems. Therefore, if you are protecting more than 10 TB of data, the DPM server must be
running on a 64-bit operating system.
See Also
Managing Performance
Performance Counters
One method you can use to monitor DPM server performance is Performance in Administrative
Tools. You can configure the monitored data to be saved as a log. You can also configure
Performance to generate alerts. For information about how to create and configure performance
alerts, see Microsoft Knowledge Base article 324752, How to create and configure performance
alerts in Windows Server 2003, (http://go.microsoft.com/fwlink/?LinkId=47881).
No
te
You can use the DPM 2010 Management Pack for System Center Operations
Manager 2007 to centrally monitor the state, health, and performance of multiple DPM
servers from an Operations Management server. To download the DPM 2010
Management Pack, see System Center Data Protection Manager 2010 Management
Pack for Operations Manager 2007 (http://go.microsoft.com/fwlink/?LinkId=189616).
The Performance Counters for Monitoring DPM table lists counters that can be useful for
monitoring DPM server performance. For more information about specific performance counters,
see Performance Logs and Alerts Help. To open the Performance tool, click Start, point to
Administrative Tools, and then click Performance. On the Action menu, click Help.
28
0
that is available to (MB). applications are
processes running on Indicates low consuming large
the specified DPM memory on DPM amounts of memory.
server. The server. • Multiple DPM jobs
Avail/MBytes value is are running
the sum of memory simultaneously.
assigned to the • The DPM server
standby (cached), does not have
free, and zero-paged sufficient memory to
lists. handle the current
DPM workload.
Processor: % Measures the > 95% for more than • Multiple DPM jobs
Processor Time percentage of time the 10 minutes. are running
processor was busy Indicates very high simultaneously.
during the sampling CPU usage on the Synchronization with
interval. DPM server. consistency check
jobs are particularly
CPU-intensive.
• On-the-wire
compression has
been enabled on the
DPM server. On-the-
wire compression
allows faster data
throughput without
negatively affecting
network
performance.
However, it places a
large processing
load on both the
protected computer
and the DPM server.
• A runaway process
is exhausting
system resources.
• The DPM server
does not have
sufficient processing
capacity to handle
the DPM workload.
Physical Disk: Current Measures the number > 80 requests for • Multiple DPM jobs
Disk Queue Length of disk requests that more than 6 minutes. that are running
(for all instances) are currently waiting Indicates possibly simultaneously are
28
1
and the requests excessive disk placing a high
currently being queue length. demand on disk
serviced. resources.
• Disk performance
needs tuning.
• Disk resources on
the DPM server are
not sufficient for the
current DPM
workload.
See Also
Managing Performance
Improving Performance
Performance is determined by workload and capacity. A slow computer might perform adequately
when it has a very light workload. In contrast, the performance of an extremely powerful computer
might suffer when challenged by an excessive workload. In operations between two computers
on a network, the workload that can be handled effectively will be limited by the component with
the least capacity, whether it is one of the computers or the network connection itself.
As a general rule, you can improve performance by making changes to the workload, the
capacity, or both.
In This Section
Modifying Workloads
Increasing Capacity
See Also
Managing Performance
Modifying Workloads
DPM offers several methods that you can use to modify protection workloads to improve
performance. The following table lists the methods you can use and indicates what you can
expect from each method.
28
2
Methods for Modifying Protection Workloads
Method Impact
Network bandwidth usage throttling Causes jobs to use less bandwidth, but they
take longer to complete.
On-the-wire compression Reduces size of data transfer but increases
CPU utilization on the DPM server and the
protected computers.
Staggering synchronization start times Balances the loads of synchronization jobs
across protection groups.
Scheduling consistency checks during off-peak Prevents DPM from interfering with regular
hours business use of protected computers.
Creating replicas manually Might make replica creation faster. There is no
performance load on the protected computer or
network resources. However, the first
consistency check will impact performance of
the protected computer.
In This Section
Using Network Bandwidth Usage Throttling
Using On-the-Wire Compression
Staggering Synchronization Start Times
Scheduling Consistency Checks
Creating Replicas Manually
See Also
Increasing Capacity
Managing Performance
28
3
Network bandwidth usage throttling is configured for each protected computer. Set network
bandwidth usage throttling in terms of an absolute maximum amount of data to be transferred per
second.
To
en
abl
e
net
wo
rk
ba
nd
wi
dth
us
ag
e
thr
ottl
ing
Network bandwidth usage can be limited by Group Policy. The Group Policy reservable
bandwidth limit on the local computer determines the combined reservable bandwidth for all
programs that use the Packet Scheduler, including DPM. The DPM network bandwidth usage
limit determines the amount of network bandwidth that DPM can consume during replica creation,
synchronization, and consistency checks. If the DPM bandwidth usage limit, either by itself or in
combination with the limits of other programs, exceeds the Group Policy reservable bandwidth
limit, the DPM bandwidth usage limit might not be applied.
For example, if a DPM computer with a 1-gigabit-per-second (Gbps) network connection has a
Group Policy reservable bandwidth limit of 20 percent, 200 Mbps of bandwidth is reserved for all
programs that use the Packet Scheduler. If DPM bandwidth usage is then set to a maximum of
150 Mbps while Internet Information Services (IIS) bandwidth usage is set to a maximum of
100 Mbps, the combined bandwidth usage limits of DPM and IIS exceed the Group Policy
reservable bandwidth limit, and the DPM limit might not be applied.
To resolve this issue, reduce the DPM setting for network bandwidth usage throttling.
28
4
See Also
Improving Performance
Modifying Workloads
To
en
abl
e
on-
the
-
wir
e
co
mp
res
sio
n
See Also
Improving Performance
Modifying Workloads
28
5
Staggering Synchronization Start Times
You can specify the starting time, in minutes after the hour, of synchronization jobs for each
protection group. Staggered starting times minimize the network impact of running multiple large
protection jobs simultaneously.
To determine whether staggering the start times of synchronization jobs is appropriate for your
needs, first gather information about scheduled protection jobs in DPM Administrator Console:
• In the Monitoring task area, on the Jobs tab, review jobs that are scheduled for times when
the DPM server experiences large disk queues.
• In the Protection task area, review details for protection groups to determine the size and
frequency of protection jobs.
Offsetting synchronization start times can also be used to optimize secondary protection of
another DPM server. Secondary protection is when a DPM server protects the database and
replicas of another DPM server, referred to as the primary DPM server. You can offset the
synchronization of the primary DPM server to the secondary DPM server to occur after the data
sources are synchronized to the primary DPM server.
To
sta
gg
er
sy
nc
hr
oni
zat
ion
sta
rt
tim
es
You can choose between two modes of synchronization: at regular intervals or just before a
recovery point is created.
28
6
Synchronization at regular intervals distributes the load on the network throughout the day. In the
case of synchronization just before a recovery point is created, the network traffic is potentially
greater at the time of synchronization, but data is not sent throughout the day.
If an organization has limited network bandwidth between the protected computer and the DPM
server and this limited bandwidth is also expected to be shared by normal corporate usage,
consider using synchronization only before recovery point and schedule it during off-peak hours.
Although the impact on network traffic and performance is important, you must also consider how
the choice of synchronization mode affects your ability to recover data. If you synchronize only
once a day, the maximum loss window is 24 hours. However, if you choose to synchronize every
hour, your maximum loss window is 1 hour.
See Also
Improving Performance
Modifying Workloads
See Also
Improving Performance
Modifying Workloads
28
7
Creating Replicas Manually
When you create a protection group, you can choose to create the replicas manually from tape or
other removable media to reduce the load on the protected computers and network.
Automatic replica creation is easier; however, depending on the size of the protected data,
manual replica creation can be faster. For smaller data sets, we recommend the automatic option.
For large data sets and slow networks, the manual option is likely to be a better choice.
After the replica is created, you must run synchronization with consistency check.
For information about how to create a replica manually, in DPM 2010 Help, see How to Manually
Create a Replica (http://go.microsoft.com/fwlink/?LinkId=197144).
See Also
Improving Performance
Modifying Workloads
Increasing Capacity
You can also improve performance by increasing the capacity of the DPM server through
hardware upgrades:
• Adding disks to the storage pool and reallocating the replicas across the storage pool can
help reduce disk queue length.
• Using striped volumes can increase disk throughput to deal with disk bottlenecks.
• Adding memory is a relatively inexpensive upgrade that can result in a noticeable
improvement in performance if the server frequently experiences low available memory.
• Adding more processors or upgrading to faster processors can reduce CPU issues.
Also, consider your data protection requirements: you might need additional DPM servers to
balance the workload.
See Also
Improving Performance
Managing Performance
Modifying Workloads
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize\Tcp1323Opts
Example: The following settings over a 100 Mbps link with 40 ms latency, gives the following
results.
Settings
On the remote DPM server: 524288
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize
On both the DPM servers: 3
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize\Tcp13
23Opts
Results
One job running 3.45 MB/sec
Three jobs running ~3 MB/sec/jo
b
28
9
• Synchronization • Copy data to tape
• Create recovery point on disk • Back up to tape
• Recovery from disk • Recovery from tape
The following table lists how protection group changes can cause the cancelation of active jobs.
Jobs can be canceled for:
• All members of the protection group ("protection group")
• All data sources on the protected computer ("protected computer")
• All protected computers in the same time zone as the computer hosting the data sources in
the protection group that is changed ("time zone")
Protection group changes and active jobs
See Also
Managing Performance
Managing Tapes
Magnetic tape and similar storage media offer an inexpensive and portable form of data
protection that is particularly useful for long-term storage.
In DPM, you can back up data from a computer directly to tape. You can also back up data from
the disk-based replica. The advantage of creating your long-term backup on tape from the disk-
based replica is that the backup operation can occur at any time with no impact on the computer
being protected.
Additionally, a thorough disaster recovery plan will include offsite storage of critical information;
you want to be able to recover your organization's data, in a situation where your facility might be
damaged or destroyed. Tape is a popular medium for offsite storage.
For additional information on DPM tape management, in the DPM Operations Guide, see
Managing Tape (http://go.microsoft.com/fwlink/?LinkId=179397).
In This Section
Short Erase
How DPM Uses Stand-Alone Tape Drives
How DPM Uses Tape Libraries
Short Erase
By default, when you erase a tape by using System Center Data Protection Manager
(DPM) 2010, DPM performs a long erase. If your tape drive supports short erase, you can use
DPM to enable it to perform a short erase by following the instructions in this topic.
29
1
Enabling Short Erase
If your tape drive supports short erase, you can enable it on the DPM server by creating the
DWORD UseShortErase under HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft Data
Protection Manager\Agent.
Im
po
rta
nt
Though short erase is much faster than long erase, it does not completely erase the data
from the tape. If you have a policy that all data from the tape must be erased and
unrecoverable, do not enable short erase.
29
2
Short-Term Tape Protection
The following table shows how the backup mode influences the number of tapes required for
short-term protection.
Example: If a full backup is scheduled weekly and incremental backups are scheduled daily, then
the first full backup will go to a new free tape and all subsequent incremental backups for six days
will be appended to another new free tape.
If a full backup job fails before it is completed, all the subsequent incremental jobs will use the
existing tape that has valid previous incremental backups.
No
te
If the customer manually triggers two individual “create recovery point (tape)” actions for
two protection group members, DPM will create two tape backup jobs and will need two
tapes to store tape backup. However, if two protection group members are selected
(multi-select in Protection view) and “create recovery point (tape)” is triggered, DPM will
use a single tape. This is designed to co-locate the data for selected protection group
members for ad-hoc tape backups onto the same tape.
29
3
No
te
Available free tapes will be decremented as tapes are allocated to either short-term or
long-term tape jobs. However, for short-term protection, creating a new recovery point will
succeed even when the "available free tapes = 0" because DPM will append the backup
job to the tape that is currently in use. Only long-term tape backups require a new tape
each time, and these backups will issue an alert if no tapes are available.
More Information
You cannot free or erase a tape that contains valid recovery points from any protected source.
Before you can free a tape, you must perform one of the following steps:
• Remove the sources from the protection group and choose to expire recovery points on the
tape.
• Change the protection group’s options and clear the tape protection options. Then, under
Inactive protection for previously protected data, right-click each data source and select
Remove inactive protection.
To restore data from an expired tape, mark the tape as free, then unmark the tape as free, and
then recatalog the tape.
29
4
• View a list of tapes
• Stop protection of a group
• Configure tape catalog retention Either the protected computer or the protection
group in the Protection task area
• Modify disk allocation Either the protected data source or the
protection group in the Protection task area
• Configure network bandwidth usage The protected computer on the Agents tab in
throttling the Management task area
• Update, disable, enable, or uninstall a
protection agent
• Lock or unlock the tape library door The tape library or stand-alone tape drive on
• Rescan the tape library the Libraries tab in the Management task area
• Clean a tape library drive The tape drive on the Libraries tab in the
Management task area
• Run a fast or detailed inventory Any tape library, stand-alone tape drive, drive,
slot, or tape on the Libraries tab in the
Management task area
• Erase a tape A tape on the Libraries tab in the
• Mark a tape as free Management task area
29
5
Classes Added by DPM
DPM adds one class, ms-SrvShareMapping, to the Active Directory directory service when you
enable end-user recovery. This class contains the mapping from the protected computer (and
share) to the DPM server (and share).
Ca
uti
on
Attribute Value
objectClass Top
objectClass classSchema
instanceType 4
possSuperiors Container
possSuperiors organizationalUnit
subClassOf Top
governsID 1.2.840.113556.1.6.33.1.22
mustContain ms-backupSrvShare
mustContain ms-productionSrvShare
rDNAttID Cn
showInAdvancedViewOnly TRUE
adminDisplayName ms-SrvShareMapping
lDAPDisplayName ms-SrvShareMapping
adminDescription Maps servers with shared resources.
objectClassCategory 1
Attribute Description
ms-BackupSrv-Share Attribute Provides the DPM share name and DPM
computer name in a string.
ms-ProductionSrv-Share Attribute Provides the protected computer share name
and protected computer computer name in a
29
6
string.
ms-BackupSrv-Share Attribute
The following table provides a detailed description of the ms-BackupSrv-Share attribute:
Attribute Value
objectClass Top
objectClass attributeSchema
attributeID 1.2.840.113556.1.6.33.2.23
attributeSyntax 2.5.5.12
rangeUpper 260
isSingleValued TRUE
showInAdvancedViewOnly TRUE
adminDisplayName ms-BackupSrv-Share
adminDescription Identifies a server with shared resources.
oMSyntax 64
IDAPDisplayName ms-backupSrvShare
objectCategory CN=Attribute-Schema,<SchemaContainerDN>
ms-ProductionSrv-Share Attribute
The following table provides a detailed description of the ms-ProductionSrv-Share attribute:
Attribute Value
objectClass Top
objectClass attributeSchema
attributeID 1.2.840.113556.1.6.33.2.24
attributeSyntax 2.5.5.12
rangeUpper 260
isSingleValued TRUE
showInAdvancedViewOnly TRUE
adminDisplayName ms-ProductionSrv-Share
adminDescription Identifies a computer with shared resources.
oMSyntax 64
IDAPDisplayName ms-productionSrvShare
objectCategory CN=Attribute-Schema,<SchemaContainerDN>
29
7
Appendix C: Custom Report Views
Data Protection Manager 2010 includes several SQL views to help you create custom reports.
SQL views simplify your queries by populating columns with data collected from multiple tables in
the database. These views offer several advantages over querying the tables directly:
• You do not need in-depth knowledge of the entire database or the relationship between tables
and keys.
• If the database structure changes in future versions of the product, the views can be updated
so that they behave the same.
For DPM installations that use a separate, dedicated computer for the SQL Server database, the
views are queried on the database computer, not the computer running DPM. This results in less
competition for resources when large numbers of views are queried over a short period of time.
The potential disadvantages of the SQL views include the following:
• Because the view runs each time it is queried, server performance may be degraded if the
view is used too frequently.
• The available supported views might not include all of the columns you need.
This appendix lists the views available in DPM.
Vw_DPM_Agents: Contains the list of computers on which a DPM protection agent from this
DPM server has been installed.
29
8
Type Integer The type of the alert
See "Alert Types" in this
appendix
Vw_DPM_CurrentOnlineMedia: The tapes that are online in DPM owned libraries currently, as
of the last inventory.
29
9
of the data source
ShadowCopyUsed Big integer The part of
ShadowCopyAllocated that is
actually in use
StartDateTime Date and time The time this statistic was
collected
EndDateTime Date and time Internal field
ScheduleType Integer The schedule period which this
0=Weekly data represents
1=Monthly
2=Quarterly
3=Yearly
Vw_DPM_DiskRecoveryPoints: Counts for disk recovery points available for each data source.
30
0
RecoverySource Integer The recovery source
0=Disk
1=Tape
30
1
AllocatedSize Big integer Disk space allocated to the
data source
UsedSize Big integer Disk space currently used by
the data source
ProductionServerName String The name of the computer on
which the data source exists
StorageNode String Always set to the DPM server
30
2
Status Integer Status of the recovery job
0/1=Progress
2=Succeeded
3=Failure
30
3
Field Data type Description
ReplicaId GUID Unique identifierentifier
generated by DPM for the
replica volume
PhysicalPath String The name of the data source on
the replica
ServerName String Name of the server to which the
data source belongs
ValidFrom Date and time When the replica was created
ValidTo Date and time The date on which the replica
was made inactive
PGId GUID Unique identifierentifier
generated by DPM for the
protection group to which the
data source belongs
StorageNode String Always set to the DPM server
Vw_DPM_TapeRecoveryPoints: Counts for tape recovery points available for each data source.
30
4
recovery points
Term Integer The schedule to which this
0=ShortTerm recovery point corresponds
1=LongTerm
30
5
1=Monthly
2=Quarterly
3=Yearly
DiskSpaceCapacity Big integer The total storage in storage
pool at end-time
PreviousDiskSpaceCapacity Big integer Total storage in storage pool
in previous corresponding
period
DiskSpaceAllocated Big integer The disk space from storage
pool that has been allocated
PreviousDiskSpaceAllocated Big integer The disk space from storage
pool that was allocated in the
previous corresponding
period
DiskSpaceUsed Big integer The actual disk space usage
PreviousDiskSpaceUsed Big integer The used disk space in the
previous corresponding
period
Alert Types
-1 RestoreDBAlert
0 NullType
1 AgentIncompatibleAlert
2 AgentUnreachableAlert
30
6
5 MediaVerificationFailedAlert
6 MediaEraseFailedAlert
7 DetailedInventoryFailedAlert
8 MediaDecommissionedAlert
9 MediaDataEraseAlert
10 FreeMediaThresholdAlert
11 DataSetCopyFailedAlert
12 BackupToTapeFailedAlert
13 BackupToTapeCatalogFailedAlert
14 LibraryDriveAlert
15 LibraryNotAvailableAlert
16 LibraryNotWorkingEfficientlyAlert
17 MediaRequiredAlert
18 ReplicaInitializationInProgressAlert
19 SynchronizationFailedAlert
20 StopProtectionFailedAlert
21 RecoveryInProgressAlert
22 RecoveryPartiallySuccessfulAlert
23 RecoverySuccessfulAlert
24 RecoveryFailedAlert
25 ShadowCopyFailedAlert
26 ReplicaInMissingStateAlert
27 ReplicaInInvalidStateAlert
28 PartialDeployedClusterAlert
29 AgentTaskFailAlert
30 SqmOptInAlert
31 DiskThresholdCrossedAlert
32 VerificationInProgressAlert
33 DiskMissingAlert
34 CatalogThresholdCrossedAlert
35 DatasetDataVerificationFailed
36 SCDiskThresholdCrossedAlert
37 ConfigureProtectionFailedAlert
38 ReplicaManualLoadPendingAlert
39 ReplicaInitializationPendingAlert
40 CertificateExpiringAlert
41 EvalShareInquiryAlert
42 ShadowCopyConsolidationRequired
30
7
Appendix D: Configuration of Inbox Tracing
The following table lists the configuration of the inbox tracing in the registry.
You can edit these values using the Registry Editor.
Wa
rni
ng
Editing the registry can produce unexpected results. Use extreme caution when changing
registry values.
Inbox Tracing
Value Name Value Allowed Values Notes/Remarks
Type
FlushIntervalInSeconds DWOR Any value Set the intervals at which
D greater than 10 inbox logs are flushed.
seconds.
TraceLogLevel DWOR Valid bitmask of Can be overridden per
D allowed trace binary.
levels.
enum TRACE_FLAG{
TRACE_ERROR =
0x2,
TRACE_DBG_ACT
IVITY = 0x4,
TRACE_DBG_
= 0x8,
TRACE_PERF =
0x20,
TRACE_DBG_FAT
AL = 0x200,
TRACE_DBG_CRI
TICAL = 0x400
};
30
8
D
<binaryname>TraceLogMaxSize DWOR Can be This is size in bytes.
D overridden to Applies to specified
change the process in process’s
default size used context. E.g.
by DPM tracing MSDPMTraceLogMaxSize
infrastructure in
the context of
individual binary.
Refer below for
binary name
mapping.
<binaryname>TraceLogMaxNumber DWOR Specify this to TraceLogMaxNumber =
D override log file [TraceLogMaxSizeInMB] /
size of individual [SingleTraceLogMaxSizeIn
log file. MB]
For example, 1 GB max
size with 15 MB each log
file size this will come out
to be 68. (1024/15 = 68)
<binaryname>TraceCriticalLogMaxNu DWOR Default = 2. Can Saves the current and last
mber D be overridden to error log separately with
save more files. extension .crash
On a critical error
logged by any
component, logs
will be saved.
Custom Actions
For a list of custom actions that are checked during DPM Setup, see DPM Setup Custom Action
Details (http://go.microsoft.com/fwlink/?LinkId=189203). These checks are part of the DPM 2010
certification for the Windows Server logo program.
The following table lists the custom actions that are performed for the DPM Installer (.msi) files.
31
0
Msdpm.msi Upgrade for the DPMRADCOMUsers group.
Dpmra.msi
Msdpm.msi Increase the IRP Stack Size.
Dpmra.msi
Msdpm.msi Add the Library agent configuration.
Msdpm.msi Add the Library authorized computer.
Msdpm.msi Install the DPMAC authorization configuration.
Msdpm.msi Disable the RSM service.
Msdpm.msi InstallDPMFilter.
Dpmra.msi
Msdpm.msi SetRebootRequired.
Dpmra.msi
Dpmra.msi AddDcomLaunchPermissions for Cmdlets.
Dpmra.msi Configure Windows Firewall.
Dpmra.msi Add the DPM server to the DCOM users group.
Dpmra.msi Add the DCOM launch permissions for RA.
Dpmra.msi Add ACL to the MTAshare.
DpmUi.msi Detect PowerShell.
DpmUi.msi Detect Windows Server 2003 R2.
SqlPrep.msi Detect DPM.
Dpmra.msi
SqlPrep.msi Detect CLI.
The following is a list of third-party binaries installed by DPM that do not have publisher
information for test case 2.8.2 for DPM certification for the Windows Server logo program.
Third-party binaries
c:\Program Files\Common Files\Microsoft Shared\DW\DWDCW20.dll
c:\Program Files\Microsoft DPM\DPM\Setup\PidGen.dll
c:\Program Files\Microsoft
DPM\SQL\90\DTS\Binn\Microsoft.SqlServer.ForEachFileEnumeratorWrap.dll
c:\Program Files\Microsoft
DPM\SQL\90\DTS\Binn\Microsoft.SQLServer.msxml6_interop.dll
c:\Program Files\Microsoft
DPM\SQL\90\DTS\Binn\Microsoft.SqlServer.SQLTaskConnectionsWrap.dll
c:\Program Files\Microsoft DPM\SQL\MSSQL.1\MSSQL\Binn\sqlmap90.dll
c:\Program Files\Microsoft SQL Server\90\Shared\sqlwvss_xp.dll
c:\Program Files (x86)\Common Files\Microsoft
Shared\MSDesigners8\msddsp.dll
c:\Program Files (x86)\Common Files\Microsoft
Shared\OFFICE11\UCS20.DLL
c:\Program Files (x86)\Microsoft DPM\SQL\90\DTS\Binn\interop.msdasc.dll
311
c:\Program Files (x86)\Microsoft
DPM\SQL\90\DTS\Binn\Microsoft.SqlServer.ForEachFileEnumeratorWrap.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\DTS\Binn\Microsoft.SqlServer.SQLTaskConnectionsWrap.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\DdsShapesLib.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\Interop.DPDPL_7_0.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\Interop.MergeModule_2_0.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\interop.msdasc.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\Interop.MSI_2_0.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\Interop.SHDocVw.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\Interop.Vdt70.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\Interop.VisioGraph_2_100.dll
c:\Program Files (x86)\Microsoft
DPM\SQL\90\Tools\Binn\VSShell\Common7\IDE\sqlresolver.dll
c:\Program Files (x86)\Microsoft Office\OFFICE11\UCSCRIBE.dll
31
2