Exchange Server 2003, even without its scheduled maintenance routines, is a very efficient messaging system. However, as mailboxes and public folders are used, there is still logical corruption. Natural wear and tear occurs, as it does in any other system. For this reason, it is important to implement a maintenance plan and schedule to minimize the impact that corruption to these databases has on the overall messaging system.
Scheduled tasks need to be performed daily, monthly, and quarterly. These recommended best practices also are intended to keep administrators informed of the status of the Exchange Server 2003 messaging environment. They can save an abundant amount of time in the long run by minimizing or even avoiding issues that can grow into bigger problems.
TIP
Document the Exchange Server 2003 messaging environment configuration and create a change log to document changes and maintenance procedures.
Daily maintenance routines require the most frequent attention of an Exchange administrator. However, these tasks should not take a significant amount of time to perform.
Daily, online backups should verify that the previous night's backup was successful. The actual verification process depends on the backup solution that is being used. In general, review the backup program's log file to determine whether the backup has successfully completed. If there are errors reported or the backup job set does not complete successfully, identify the cause of the error and take the appropriate steps to resolve the problem.
In addition, it is also a best practice to do the following to back up an Exchange Server 2003 server:
Include System State data to protect against system failure.
Keep note of how long the backup process is taking to complete. This time should match any service level agreements that may be in place.
Verify that transaction logs are deleted if circular logging is disabled. If not, perform a full backup.
All volumes that Exchange Server 2003 resides on (Exchange system files, databases, transaction logs, and so forth) should be checked on a daily basis to ensure that ample free space is available. If the volume or partition runs out of disk space, no more information can be written to the disk. Without disk space to write to, the Exchange services stop running.
Message queues should be checked daily to ensure that there are no messages stuck in the queue. Use Queue Viewer to view and manage SMTP, MTA, and connector messaging queues to keep messages flowing.
On Exchange Server 2003 servers, check the Application Log within the Event Viewer for Warning and Stop error messages. These error messages might directly lead you to an issue on the server, or some error messages may be symptomatic of other issues. Filtering for these event types can save a lot of time evaluating whether one of these events has occurred within the last 24 hours. If you are using a systems or operational management solution, this process and more can be automated. In addition, these solutions can also provide enhanced reporting functionality.
Tasks that do not require daily maintenance, but still require frequent attention, are categorized in the weekly maintenance routines. These routines are described in the following sections.
Unless you set mailbox storage limitations, the size of the mailbox databases can quickly become overwhelmingly large. If the volume housing the databases is not large enough to accommodate the database growth beyond a certain capacity, services can stop, databases can get corrupted, performance can get sluggish, or the system can halt. Even when you do set mailbox size limitations, you should know the size of the databases and the growth rate. By documenting the database size(s), you can better understand system usage and capacity requirements.
Many environments rely on public folders to share information, and the public folder configurations (for example, multiple hierarchies, multiple replicas, and more) vary widely from environment to environment. With environments that replicate public folder information among different Exchange Server 2003 servers, keep abreast of whether the information contained within those folders is kept up to date.
There are several ways to perform quick tests to see whether information is replicating correctly, including manually testing replication, using the Exchange Server 2003 Resource Kit's Public Folder Administration tool (PFAdmin.exe), and reviewing the Ex00yymmdd.log and Ex01yymmdd.log files. If problems exist, you can use the logs just mentioned to troubleshoot.
Exchange Server 2003 records information in the Application Log about online maintenance that occurs automatically. Check this event log to verify that all the online maintenance tasks and other scheduled tasks are being performed and that no problems are occurring. Using the common event IDs given in the following list, you can easily search by the ID number to review online maintenance and other scheduled tasks. In the right pane of the Event Viewer, click on the Event column to sort events by their ID number:
Event ID 1221 This event reveals how much free space there is in a database. This information is also useful in determining when offline database defragmentation may be necessary.
Event ID 1206 and 1207 Both IDs inform about deleted item retention processing.
Event ID 700 and 701 These IDs indicate the start and stop times of the online database defragmentation process. Check to make sure that the process does not conflict with Exchange database backups, and make sure the process completes without interruptions.
Event IDs 9531–9535 All these IDs are about deleted mailbox retention processing.
With any system—and Exchange Server 2003 is no different—it is important to know how well the overall system is performing. At a minimum, you should monitor system resources at least once a week. Concentrate on monitoring the four common contributors to bottlenecks: memory, processor, disk subsystem, and network subsystem.
An Offline Address Book (OAB), also known as an Offline Address List (OAL), is routinely generated for remote users to download and view address lists while offline. By default, the OAL is generated daily if there are changes. Use the ESM to determine the last time it was generated to make sure that remote users can obtain an updated copy. This is performed by viewing the Property pages of the OAL located under the Recipients, Offline Address List container.
NOTE
If you are experiencing problems with OAL generation, enable diagnostic logging and review the Application Log for any OAL Generator category events.
Recommended monthly maintenance practices for Exchange Server 2003 do not require the frequency of daily or weekly tasks, but they are nonetheless important to maintaining the overall health of the system. Some general monthly maintenance tasks can be quickly summarized; others are explained in more detail in the following sections.
General tasks include
Refresh the Exchange Server 2003 services to free up memory resources and kick-start online maintenance routines.
Install approved and tested service packs and updates.
Schedule and perform, as necessary, any major server configuration changes, including hardware upgrades.
Uninterruptible Power Supply (UPS) equipment is commonly used to protect the server from sudden loss of power. Most UPS solutions include supporting management software to assure that the server is gracefully shut down in the event of power failure, thus preserving the integrity of the system. Each manufacturer has a specific recommendation for testing, and its procedure should be followed. However, it should occur no less than once a month, and it is advantageous to schedule the test for the same time as the server reboot.
As mentioned earlier in “Performing Offline Database Maintenance,” an approximation of a database's fragmentation can be made using the database size and the amount of free space. The amount of free space that can be recovered from a defragmentation and compaction is provided within Event ID 1221 entries.
Although quarterly maintenance tasks are infrequent, some might require downtime and are more likely to cause serious problems with Exchange Server 2003 if not properly planned or maintained. Therefore proceed cautiously with these tasks.
General quarterly maintenance tasks include the following:
Check mailbox and public folder stores' Property pages to verify configuration parameters, review usage statistics, determine mailbox sizes, and more.
Check storage limits to ensure that data storage requirements will not exceed capacity, given the current rate of growth (stemming from the information taken from the weekly maintenance task).
As mentioned earlier, offline maintenance is one of the most important maintenance tasks to perform, but it can be time consuming and hazardous. Remember to properly plan, schedule during off-hours, and perform both an online and offline backup of the information stores prior to beginning the tasks. A little extra care up front can save you lots of time troubleshooting. For more information on this process, refer to “Performing Offline Database Maintenance,” earlier in this chapter.
At first glance, you might consider the process of validating database backups as simply checking the backup logs to see whether they were successful. On the contrary, validating the backups involves performing a full restore onto a test server in a lab environment. This not only ensures that Exchange Server 2003 can be easily recovered in times of disaster but it also irons out any issues in the restore process and keeps administrators in practice for recovering the system; when disaster strikes, they are adequately prepared.
TIP
Document the process of restoring Exchange Server 2003 databases. If documentation already exists, verify that the existing process has not changed. If it has changed, update the documentation.