Tips » Disaster Recovery and Backup » Designing for Disaster Recovery with Exchange

Designing for Disaster Recovery with Exchange

by Eric Beehler

With all the buzz around high-availability for Exchange 2007, some have wondered whether a discussion of disaster recovery is even necessary and if backups are obsolete—ancient history for Exchange. It is true that you can utilize the high-availability options in Exchange 2007, such as Cluster Continuous Replication (CCR), combined with Standby Continuous Replication (SCR) to provide such a solid high-availability solution that you could almost say you are ready for a disaster.

However, there are many reasons that setup might not be possible for your particular situation. Your environment might not truly need several Exchange servers (which you would need in order to establish CCR) nor a reason for a secondary site and location (which you would want if you were going to make full use of SCR) nor the budget for the hardware and software that is required to implement such solutions (considering multiple servers and the fact that you need to run Enterprise Edition of Windows Server to implement the CCR cluster). Disaster recovery solutions may be what your environment should design. But what does that entail?

Assuming that the three primary recovery points are messages, mailboxes, and/or databases (within storage groups), let’s consider some of the options you have at your disposal for recovering from a disaster. Keep in mind that a disaster might be of the natural or man-made sort or an accidental occurrence such as a database corruption, a disk failure, or a server failure.

Disaster Recovery Solutions

Typically, the primary server to be concerned about amongst your server roles is the Mailbox Server, so that will be our focus. However, there are non-Mailbox Server roles that you would also want to consider for restoration in the event of a disaster. For the purposes of this discussion, we will consider the following items:

  • Backup/recovery solutions
  • VSS backups
  • Database portability
  • Recovery storage groups
  • Dial-tone recovery

Backup/Recovery Solutions

In terms of backing up your Exchange databases and mailboxes, with legacy Exchange implementations (Exchange 5.5, 2000, and 2003), your built-in backup/recovery tools within the Windows Server OS (NTBackup) was enhanced to allow for the backup of your Exchange storage groups. Such continues to be the case if you have installed Exchange 2007 on Windows Server 2003. You can perform a streaming backup (not a VSS backup) of the local system. However, Server 2008 does not (at the time of this writing) support a streaming online backup of the Exchange databases. Thus, you will have to pursue a third-party solution.

Note: The Microsoft Exchange team has mentioned that a future fix will correct this shortcoming. While you wait, if you want to use Windows Server 2008, you will need to consider third-party alternatives for backup. One consideration is Microsoft’s System Center Data Protection Manager 2007.

VSS Backups

Volume Snapshot Service (typically called Volume Shadow Copies) is a solution that is part of the Windows OS; it is used already to provide a modicum of recoverability in terms of system restore and/or previous versions; however, in relation to Exchange, the functionality is built-in to be taken advantage of by Server 2003 and 2008. Be aware that this requires a third-party solution to utilize the VSS ability of taking a point-in-time snapshot of your data.

Note: There are a variety of third-party backup solutions for Exchange but not all take advantage of VSS functionality. One example is Robobak, which provides a thin-backup solution for Exchange that is agentless, designed for remote and branch office data protection, offers “incremental forever” backups, provides de-duplication for single instance of data, and has on-demand synthetic restores. This option is an excellent solution, but one that doesn’t currently use VSS, so be aware of your needs and wants and look for a solution that fits.

Database Portability

Before discussing the virtues of recovery storage groups, it is good to mention here that Exchange 2007 has simplified database portability to such a tremendous degree that you can actually restore any mailbox database to any server within the same organization. Although previous versions required you to have a server with the same name, such is not the case with Exchange 2007. If you need to relocate a database due to maintenance or disaster, you can move it from one storage group to another, from one disk to another, or from one server to another.

Recovery Storage Groups

Rather than restoring a storage group directly to a server, there are times when you might need only a mailbox or message to be restored. In these situations, you don’t have to restore your backup to a dummy server; you can restore it to a recovery storage group and then move the message, mailbox, or database (mailbox not Public Folder) over to the functioning database.

Figure 1: Creating a recovery storage group.

To accomplish this task, you need to work through your Exchange Consoles Toolbox from the Database Recovery Management Tool. You need to first create the recovery storage group (as you see in Figure 1), and then you can restore the backup to that storage group.

Dial-Tone Recovery

If your mailbox servers crashes, you need to get your users up and running fast with the ability to send and receive email. If that is your primary concern, you can perform a dial-tone recovery. Because the actual mailbox data is part of Active Directory (AD), you can restore the mailbox configuration information very quickly and then restore the mailboxes at leisure (well, not really leisure, but without users breathing down your neck about sending and receiving email). Keep in mind with the dial-tone recovery that users will open Outlook and will see no mail from the past (because that hasn’t been restored). You should warn them of this fact. Later on when the database is restored, you are going to swap the new and old databases. However, you are still going to need to recover the day’s mail and put it in the newly restored databases. This task involves the same process as restoring mail from a standard recovery. So, you restore the database to the recovery storage group, you swap the restored older version with the newer day’s mail version, and then you merge over the day’s mail with the recovered database. Sounds a bit complicated but it’s just a matter of following the proper steps.

Don’t Panic…Plan Ahead

The key to disaster recovery is to know your options and then set about planning for the worst. Planning involves ensuring you have everything documented (and the documentation is held offsite), everything backed up or duplicated in one way or another for recovery (and again, offsite, and if you are really worried, out-of-state or country), and you run practice drills to ensure you know exactly what to do and how.

Eric Beehler has been working in the IT industry since the mid-90’s and has been playing with computer technology well before that. From Help desk technician to solutions provider, he has been involved at many layers of enterprise solutions from the desktop to the network to the server and the SAN. He currently has certifications from CompTIA (A+, N+, Server+), and Microsoft (MCITP: Enterprise Support Technician and Consumer Support Technician, MCTS: Windows Vista Configuration, MCDBA SQL Server 2000, MCSE+I Windows NT 4.0, MCSE Windows 2000, and MCSE Windows 2003). He also holds a Master’s degree in Business Administration from the University of Colorado at Colorado Springs. His experience includes more than nine years with Hewlett-Packard’s Managed Services division, working with Fortune 500 companies to deliver network and server solutions and, most recently, IT experience in the insurance industry working on highly available solutions and disaster recovery. He has co-authored books, including MCITP: Microsoft Windows Vista Desktop Support Enterprise Study Guide (Sybex/Wiley Publishing), authored several white papers, and co-hosts the "CS Techcast" podcast aimed at IT professionals. He provides consulting and training through Consortio Services, LLC.