News Stay informed about the latest enterprise technology news and product updates.

Best Practice #1: Understanding Exchange databases

Best Practices Checklist: Exchange Server disaster recovery planning , Best Practice #1: Understanding Exchange databases.

The Extensible Storage Engine (ESE) and its associated files are probably the least understood components of Exchange for many administrators. This can unfortunately lead to a number of misconceptions and bad backup and restore practices.

Here are a few basic database concepts all administrators should understand before creating a disaster recovery plan:

  • Database: In Exchange 2000/2003, the database is comprised of two files -- an .edb (rich text data) and .stm (native content) files. Databases can be either mailbox stores or public folder stores.

  • Transaction logs: Before data is written to the database, it is first written to a sequential set of transaction logs. These logs are not only essential to the performance of Exchange, but they are also a key ingredient in an Exchange disaster recovery plan.

  • Consistent database: When a database is dismounted, or the information store service (store.exe) is shut down, all logs are committed to the database. After all logs have been committed, the header information in the database is flagged as "consistent."

  • Inconsistent database: When a database is stopped prior to all transactions being committed (because of a power outage, for example), the header of the database is not flagged and is therefore regarded as "inconsistent." To make the database consistent, the transactions that were not committed must be replayed into the database. A database must be consistent in order for it to be mounted.

Best Practices Checklist: Exchange Server disaster recovery planning

 Home: Introduction
 Best Practice #1: Understanding Exchange databases
 Best Practice #2: Building your plan around the technology at hand
 Best Practice #3: Keeping e-mail in perspective
 Best Practice #4: Configuring server hardware for disaster recovery
 Best Practice #5: Configuring Exchange for disaster recovery
 Best Practice #6: Simulating a disaster
 Best Practice #7: Learning from others' mistakes and successes
 Best Practice #8: Considering offsite storage and remote recovery
 Best Practice #9: Familiarizing yourself with the right resources

Richard Luckett, Vice President and Senior Consultant, Ajettix Security
Richard Luckett is a Microsoft Certified Systems Engineer on the Windows NT 4.0, 2000 and 2003 platforms and has been certified on Exchange since version 4.0. He is the co-author of Administering Exchange 2000 Server, published by McGraw Hill, and has written four Exchange courses, Introduction to Exchange 2000, and Hands-on Exchange 2003, Ultimate Exchange Server 2003 and Exchange Server 2003 Administrator Boot Camp for Global Knowledge Inc. Richard is currently Vice President and Senior Consultant for Ajettix Security, where he is the head of the Microsoft security practice.

Dig Deeper on Exchange Server setup and troubleshooting

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.