Choose a backup strategy that lets you meet your business requirements and service level agreements. For example, allowed downtime, allowed recovery time, and data loss tolerance.
A backup strategy includes the following:
Choose backup hardware.
Choose backup application.
Identify the things that you want to back up.
Choose a strategy for recovering an entire server. For example, restoring the server, rebuilding the server, or using a stand-by recovery server.
Determine a backup schedule for each category of data that you want to help protect and the frequency of backup types (normal, incremental, differential) for each category.
Ensure that You Help Protect the Data You Need
Identify the components that can be restored by replicating data from other sources (for example, data that is stored in Active Directory directory service), what components must be restored from backups (such as Exchange Server databases and transaction log files), and what data can be recreated (such as connector and server configuration).
The data you help protect can include the following:
Windows operating system data
Domain controller data
Internet Information Services (IIS) metabase
Exchange Server 2003 databases and transaction logs
Cluster configuration data (if you are using back-end clustering)
Individual mailboxes (optional)
Unique dynamic data: preserve any other data stored on your servers particular to your organization that would be difficult to re-create or restore. For example, this data might include custom scripts, Web forms for IIS, or any other data that you want to help protect.
Implement Practices that Minimize the Effect of a Disaster
Consider implementing the following measures to help prevent or minimize the effect of a disaster:
Have your software and firmware updates available.
Have all software disks readily available.
Have a plan to monitor servers proactively.
Maintain hardware records.
Maintain software records.
Implement fault tolerance in your organization at the hardware or software level. This would include using methods like RAID, multi-path hardware solutions, and clustering.
Ensure that you have sufficient hard disk capacity for your servers running Exchange Server. You should have sufficient space on your hard disk to restore both the database and the log files. For more information about disk capacity planning, see the Hard Disk Space Considerations topic in Chapter 4 "Planning a Reliable Back-end Storage Solutions" in the Exchange 2003 High Availability Guide.
Use high quality media.
Plan schedules for archiving your backup media.
Archive the backup media in a secure location, for example, a fire-proof safe or at another location.
Train your staff on disaster recovery procedures.
Monitor the health of the Exchange store. For example, monitor the event log for the occurrence of 1,221 events to determine the amount of white space available in a database. If the available white space equals 30 percent of the size of the database, you might want to consider offline defragmentation of Exchange Server databases.
Document and test your recovery procedures.
Verify the integrity of your backups. Ensure that they occur without error.
Perform disaster recovery simulation drills.
Keep spare hardware available.
Put your transaction log files and database files on separate physical drives.
Distribute your users across multiple mailbox stores.
Consider creating archival servers to store mailboxes that have not been accessed for a pre-determined number of days. This keeps an inactive mailbox from using resources and possibly growing uncontrollably.
Put mission-critical mailboxes all in one database for quick restoration, if your business requires it. You can also back up the mailboxes individually (brick-level backups) with non-Microsoft backup software.
Configure deleted item retention for your users.
Configure deleted mailbox retention at the mailbox level.
Implement practices to minimize Exchange Server database restore times.
Ensure that your insurance policy is sufficient.
To increase the efficiency of defragmentation and backup processes, schedule your maintenance processes and backup operations to run at different times.
Ensure that the online maintenance process is completed successfully. For example:
Event: 1221 Source: MSExchangeIS Private Type: Information Category: General Description: The database has nnn megabytes of free space after online defragmentation has terminated.
Event: 1221 Source: MSExchangeIS Public Type: Information Category: General Description: The database has nnn megabytes of free space after online defragmentation has terminated.
Practice restoring from backup in a test environment.