Continuous Backup for Exchange Server 2010 with DPM 2012 (Part 4)

If you would ike to read the first part in this article series please go to:

Recovering Exchange 2010 Mailbox Database

DPM 2012 supports recovery of the following Exchange data:

  • An entire storage group (Exchange 2007 or lower).
  • A single mailbox database.
  • A single mailbox.

Depending on the type of information to recover, you also have some flexible restore options where to dump the data:

  • To its original location.
  • To a Recovery Storage Group / Recovery Database.
  • To a separate network folder.
  • To a tape drive.

Besides all the discrete recovery points that are available (and they can be every 15 minutes), DPM provides one more recovery capability: to recover “latest”, resulting in near zero data loss when the recovery is complete. In the event of a complete loss of the Exchange database files, one can choose “latest” which instructs DPM to first restore to the last recovery point and then roll forward the surviving transaction logs beyond that, as long as the best practice of storing Exchange databases on one volume and logs on another has been respected.

For this first recovery scenario, we’ll recover the Contoso-Users-01 database using the “latest” capability. Let’s take a look at the contents of the James Bond mailbox before our simulated disaster happens (Figure 1).

Next, to simulate the complete loss of the Contoso-Users-01 database, I stopped the Microsoft Exchange Information Store service and renamed the .edb file, which caused it to appear as dismounted in the Exchange Management Console (Figure 2). Please also notice the warning I got after trying to start Microsoft Exchange Information Store service (Figure 3).


Figure
1: James Bond mailbox


Figure
2: Contoso-Users-01 database failed


Figure
3: Contoso-Users-01 database missing

To recover a Mailbox Database to its original location, follow these steps:

  1. Open DPM Administrator Console and click Recovery on the navigation bar. Browse to the mailbox database you wish to recover in the All Protected Exchange Data node. Click any bold date in the calendar, select the Latest recovery point from the Recovery time drop-down box and click Recover in the Actions pane to launch the Recovery Wizard (Figure 4).


Figure 4: Selecting a mailbox database recovery point

  1. Review the recovery selection and click Next (Figure 5). Select Recover to original Exchange Server location and click Next (Figure 6).


Figure
5: Review Recovery Selection


Figure
6: Select Recovery Type

  1. On the Specific Recovery Options page, make sure Mount the databases after they are recovered is selected. If you want DPM to send a notification when the recovery process is finished, select the Send an e-mail when this recovery completes check box and enter one or more e-mail addresses (Figure 7). Click Next.


Figure
7: Specific Recovery Options

  1. Now we must go back to the Exchange Server in order to allow the databases to be overwritten by the restore. If you miss this step, the restore will fail. Open your Exchange Management Console, navigate to Organization Configuration > Mailbox, right-click the desired Exchange mailbox database and select Properties. Select the This database can be overwritten by a restore check box and click OK (Figure 8).
    Back to the DPM server, on the Summary page, review your selected settings and click Recover (Figure 9). When the recovery is complete, click Close (Figure 10).


Figure
8: Contoso-Users-01 Properties


Figure
9: Summary


Figure
10: Recovery Status

We can now check the James Bond mailbox contents in order to verify the success of the restore operation (Figure 11).


Figure
11: James Bond mailbox after a successful restore of the Contoso-Users-01 database


Figure 12:
DAG database copy failed

Since the mailbox database recovered was part of a DAG, the passive copy shows the failed and suspended status. In order to resume normal DAG operations, right click the failed database copy and select Resume Database Copy (Figure 13). You’ll get a pop-up window warning that the database needs to be reseeded (Figure 14). Click Yes.


Figure
13: Resume Database Copy


Figure
14: Resume Mailbox Database Copy

Recovering Individual Exchange 2010 Mailbox

DPM 2012 supports the recovery of individual mailboxes, although unlike other products that do brick-level backups, 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 support policy for third-party products that modify or extract Exchange database contents“.

Individual mailbox recover is done to a recovery database rather than directly to the database that hosted the original mailbox. The recovery database must exist prior to attempting this recovery

If you follow some best practices, such as keeping deleted items and deleted mailboxes for a period of time, the probability that you’ll ever need to restore a single mailbox is quite low. Nevertheless if you cannot recover the needed data using these methods, here are the steps to recover a protected mailbox using DPM 2012:

  1. On the protected Exchange server, if you do not have an existing Recovery Mailbox Database, create one by using the New-MailboxDatabase cmdlet in Exchange Management Shell. Configure the recovery database to allow it to be overwritten by using the Set-MailboxDatabase cmdlet in Exchange Management Shell.
    a)  New-MailboxDatabase -Recovery -Name RDB-Contoso-Users-01 -Server E2K10-MBX2
    b)  Set-MailboxDatabase -Identity “RDB-Contoso-Users-01” -AllowFileRestore $true


Figure
15: New-MailboxDatabase


Figure 16:
RDB-Contoso-Users-01

  1. Open DPM Administrator Console and click Recovery on the navigation bar. Navigate to the mailbox database you wish to recover, under All Protected Exchange Data pane, to display the list of available mailboxes. Click a date in the calendar, select a recovery point from the Recovery time drop down box and click Recover in the Actions pane to launch the Recovery Wizard (Figure 17). When you select a mailbox for recovery, you cannot select Latest as the recovery point, this functionality is not available for individual mailboxes (Figure 18).


Figure 17: Selecting a mailbox recovery point


Figure
18: Select Recovery Type

  1. Review the recovery selection and click Next (Figure 19). Select Recover mailbox to an Exchange server database to recover the mailbox to its original server, or select Copy to a network folder to copy the database files to a separate folder location. Click Next (Figure 20).


Figure
19: Review Recovery Options


Figure
20: Select Recovery Type

  1. On the Specify Destination window, specify the location to write the recovered mailbox database to (The Recovery Mailbox created in step 1), and click Next (Figure 21).


Figure
21: Specify Destination

  1. On the Specify Recovery Options, choose whether you want to use Network bandwidth usage throttling and SAN based recovery using hardware snapshots if available and applicable. Specify if you want DPM to send an e-mail message when the recovery process is finished and click Next (Figure 22).
  2. On the Summary page, review your selected settings and click Recover (Figure 23). When the recovery is complete, click Close (Figure 24).


Figure
22: Specify Recovery Options


Figure
23: Summary


Figure
24: Recovery Status

  1. After the recovery process finishes, we still don’t have the required mailbox restored. What we have is the mailbox database where the mailbox belongs to restored to the Recovery Mailbox. On the protected Exchange server, we can list all the mailbox databases by running the GetMailboxDatabase cmdlet (Figure 25).


Figure
25: Get-MailboxDatabase

  1. The final step to restore the mailbox is to run a PowerShell cmdlet:
    a)  New-MailboxRestoreRequest -SourceDatabase “RDB-Contoso-Users-01” -SourceStoreMailbox “James Bond” -TargetMailbox [email protected] -TargetRootFolder Recovery
    If we now open the James Bond mailbox, all its contents until 3:00 PM are located beneath the Recovery folder (Figure 27).


Figure 26:
Recovery Storage Group files


Figure
27: James Bond mailbox after restore

After you complete the restore, the Recovery Mailbox can be dismounted and deleted, using the following PowerShell cmdlet:

“Remove-MailboxDatabase -Identity “RDB-Contoso-Users-01”


Figure
28: Remove-MailboxDatabase

Reporting

DPM reporting offers six standard reports that you can generate, review, and analyze. The following table lists the reports and their definitions.

Report Name

Summary of Contents

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.

Tape Management

The Tape Management report provides details for tape rotation and decommissioning, and it verifies that the free media threshold is not exceeded.

Use this report to manage tape circulation between the library and your offsite location.

Tape Utilization

The Tape Utilization report provides trending of resource (disk/tape) usage over time to assist capacity planning.

Use this report to make decisions about tape allocations and purchases.

Protection

The Protection report provides the commonly used metrics for backup success rolled up over long periods of time to track how backups are doing.

Use this report to identify which computers or protection groups have been backed up successfully.

Recovery

The Recovery report provides the commonly used 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:

¡ì   Identify trends in disk usage

¡ì   Make decisions about modifying space allocations for protection groups and adding disks to the storage pool

¡ì   Identifying how much disk resource each computer is using on DPM

Table 1: DPM Report Types

What can you do in the Reporting task area (Figure 29)?

  • Generate and view reports on DPM operations.
  • Schedule automatic report generation.
  • Manage Reporting Services settings.

Figure 30 shows an example of a Disk Utilization report.


Figure
29: Reporting pane


Figure
30: Disk Utilization report

Best Practices

Before concluding this series of articles, I’d like to highlight some best practices for operating a DPM deployment.

  • When protecting DAGs, it is not necessary to select the active mailbox databases for protection by DPM. DPM is database role agnostic and can be configured to protect a server that hosts a collection of active or passive mailbox databases.
  • You need to configure at least one Full Backup per day. The Full Backup backs up the mailbox database and log files and then truncates the log files. When protecting more than one copy of an Exchange mailbox database (such as when you are protecting multiple members of a database availability group) you should configure one node for full backups and the rest as copy backups. Copy backups do not truncate log files.
  • If Exchange is implemented with inexpensive SATA/JBOD disks, protect at least two copies of each mailbox database.
  • The DPM default setting has Exchange 2010 running ESEUTIL on Exchange mailbox databases as a background task to verify their integrity. DPM allows you to skip running ESEUTIL on the mailbox database and only use the utility to verify the integrity of the log files (recommended for DAGs).
  • If the Exchange DAG node hosting the copy of the mailbox database on which the Full Backup is taken goes down temporarily, it is not necessary to perform any steps on the nodes on which copy backups are taken. If it becomes necessary to switch to another node because the failed node will no longer be available, you need to reconfigure DPM to take a full backup of the new target node.
  • Regarding the synchronization frequency, setting the minimum period (15 minutes) would probably be overkill. Start by setting up your current backup policy, and then gradually increasing the number of recovery points. I think that 1 or 2 Express Full Backups per day, plus a synchronization frequency of 2 hours is a nice number. Of course, there’s no such thing as a one size fits all, you have to take into account the volume of your data, the impact on the performance of your servers (DPM has a reduced impact, but still it has some impact) and the volume required to store the replicas.

Conclusion

Data Protection Manager 2012 provides fully integrated data protection for Microsoft Exchange 2003, 2007 and 2010. When using DPM 2012 you can have confidence that:

  • DPM can be used to reliably and quickly back up Exchange production servers at frequent intervals without a strong impact on performance.
  • DPM replicas and recovery points allow organizations to reliably restore Exchange data not only in its most recently backed up state, but in any state that was captured in recovery points.
  • Exchange mailbox data can be restored not just to the original Exchange server, but to other protected servers and arbitrary file locations.

By using combined disk-to-disk, disk-to-tape and disk-to-disk-to-tape technologies, DPM helps to ensure that an Exchange infrastructure is better protected and always available

Related Links

If you would ike to read the first part in this article series please go to:

Leave a Comment

Your email address will not be published.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Scroll to Top