Microsoft KB Archive/128325

= XADM: Reclaiming Disk Space for the Information Store =

PSS ID Number: 128325

Article Last Modified on 4/28/2005

-

The information in this article applies to:


 * Microsoft Exchange Server 4.0

-



This article was previously published under Q128325



SUMMARY
If the Information Store does not start due to lack of disk space, an application event is logged in the Windows NT Event Viewer. The source is EDB and the error test includes the Jet Blue error ID -1808. This article describes how to free up disk space so that the Information Store can restart, and how to use Windows NT Performance Monitor to track disk space usage in the future.



Monitoring Disk Space
An Administrator can observe disk space usage on the drive containing the Information Store by using the Windows NT Performance Monitor. The LogicalDisk object along with the "% Free Space" and "Free Megabytes" counters are used to monitor and trigger alerts when disk space is low.

Recovering Space Used by Log Files
Increasing log files can cause the Information Store or Directory Store to run out of operating space. To prevent this, do one of the following:
 * Write the log files to a different drive.
 * To change the location where the Information Store or Directory Store Transaction logs are written, select the Server object properties page and choose the Database Paths tab. Change the path for the Information Store and Directory Store transaction logs and click OK.
 * Backup the Exchange Server.
 * Use the Windows NT Backup utility that ships with Exchange Server to perform either a Normal (full) or Incremental Online backup of the server. The Backup utility automatically deletes transaction logs that are no longer needed (they have been committed to disk). If you never run the Backup utility, log files continue to grow.
 * Correctly delete log files periodically.
 * When the Information and Directory stores run, you find log files of the EDB*.LOG form in both the EXCHSRVR\DSADATA and EXCHSRVR\MDBDATA directories. Microsoft Exchange uses these log files to commit data to disk. In addition, these log files are necessary to perform Incremental and Differential type backup using the Windows NT Backup utility.
 * The EDB*.LOG files can be deleted after the Directory and Information Store services have been shut down. If the log files are deleted when the services are still running, you can jeopardize data integrity. When the services are stopped, the transactions in these log files are committed to the main database file on disk (DIR.EDB for the directory and PRIV.EDB and PUB.EDB for the Information Store). When this data is committed, the log files are no longer needed. Follow the steps below to correctly delete log files:
 * In Control Panel Services, select the Microsoft System Attendant Service and click STOP. This stops all Microsoft Exchange Services.
 * Make a backup copy of the EDB*.LOG files in both the EXCHSRVR\DSADATA and the EXCHSRVR\MDBDATA directories. Copy them to an alternate location on the hard drive or to the tape backup.
 * Delete all the EDB*.LOG files in both the EXCHSRVR\DSADATA and EXCHSRVR\MDBDATA directories.
 * Restart all the Microsoft Exchange Server Services (in particular, the Directory and Information Store services).
 * If these services do not start, restore the DSADATA and MDBDATA sub-directories from tape or hard disk. Restart the services again.
 * If the services start, delete the backup copy of the DSADATA and MDBDATA sub-directories.

NOTE: If you delete the EDB*.LOG files in the manner described above, Incremental and Differential backups of the Exchange Directory Store and Information Store are no longer valid or possible. You can only perform a Normal type (full) backup.

Circular Logging (which is enabled, by default - Advanced tab of the server object's property page) writes log files, but once the checkpoint has been advanced, the inactive portion of the transaction logs are discarded. Typically, this represents the majority of the potential log data. The total size of the active transactions are less than the total amount of RAM on a given computer. Therefore, with circular logging enabled, the system has complete recoverability, with respect to hard and soft crashes. The element that is sacrificed is the protection against media failure. These methods are not supported on servers, where Circular logging is enabled, because the transaction logs are used for incremental and differential backups. An error message in Windows NT Backup appears.

Checking a Database for Consistency
A Microsoft Exchange Server database is considered consistent only after normal shut down. At all other times, including during normal operation, a flag in the database marks it as inconsistent. Thus, if the database service is terminated abnormally, Exchange Server detects something wrong on the next startup. Exchange then initiates "soft recovery" steps to back out or commit necessary transactions to the database and to restore its integrity.

You can check a database for consistency with the following command:

eseutil /mh | more

The State line in the screen output from this command will contain Consistent or Inconsistent.

You should not move inconsistent databases. Log files need to be played into that database to make it consistent. Because the path is encoded into the log files header information for the database, replay will not occur if you move the databases.

For additional information on how to resolve a full transaction log drive, click the article number below to view the article in the Microsoft Knowledge Base:

259751 XADM: How to Recover from a Full Transaction Log File Drive

Keywords: kbenv KB128325

Technology: kbExchange400 kbExchangeSearch kbZNotKeyword2

-

[mailto:TECHNET@MICROSOFT.COM Send feedback to Microsoft]

© Microsoft Corporation. All rights reserved.