8.6 Offline address lists

Chapter 9 - Backup and Recovery Operations
Monitoring and Managing Microsoft Exchange 2000 Server
by Mike Daugherty
Digital Press 2001

9.1 Minimizing risk

Hardware failures, software failures, human error, and sometimes even natural disasters can disrupt your e-mail environment. Disasters happen, and you must be prepared to respond quickly. If you use the following practices, you can reduce the risk and impact of potential disasters: Ensure that circular logging is turned off for all Storage Groups. With circular logging enabled, transaction logs are overwritten to save disk space. However, overwriting transaction logs prevents the overwritten logs from being used during recovery operations.

• Perform daily full (normal) backups of the Exchange Information Store.

• Perform periodic full backups of Windows 2000 and Exchange 2000 configuration data.

• Select server-class hardware for your servers rather than high-end desktop systems. Redundant power supplies , multiple processors, and hardware RAID are worth the extra cost.

• Install all Exchange servers in a controlled environment consistent with the manufacturers recommendations. Protect the servers with Uninterruptible Power Supplies (UPS). Physically secure the environment that houses the servers.

• Protect databases using hardware RAID-5 (disk striping plus parity) or RAID 0+1 (disk striping and mirroring) technology.

• Keep transaction log files on separate hard drives from the databases. Protect the log files using RAID-1 (disk mirroring) technology.

• Keep the Windows 2000 operating system files on a separate hard drive and protect them using RAID-1 (disk mirroring).

• Ensure that your Exchange servers have adequate disk space, including sufficient space to support recovery operations.

• Have multiple Windows 2000 domain controllers for each domain to provide redundancy in the event of a single failure. Three domain controllers are recommended. If you only have two DCs, then you are at risk whenever you take one DC off-line for maintenance. With three DCs, you are still protected if one of the DCs fails while you have one temporarily off-line for maintenance.

• Maintain up-to-date documentation for your server configurations.

• Have a dedicated recovery server with the same configuration as your production servers. A dedicated recovery system is one that is only used when a disaster occurs. The server is not connected to the network.

• Fully document your recovery procedures and regularly practice disaster recoveries .

By following these practices, you can reduce the risk and impact of disasters, but you cannot completely avoid disruptions.