Before running the Eseutil repair mode on your mailbox or public folder database, do the following:

• Check for sufficient disk space on the local logical drive for the temporary repair database. Keeping 20 percent of the size of the database files being repaired is suggested, although the size of the temporary file will vary widely depending on the nature of the repairs made. If sufficient space is unavailable, you may redirect temporary files to a different drive, as described later in this topic.
  
  
• Check that you have full permissions for the folders you want to perform the repair on.
  
  
• Make sure you log on by using an account that is delegated membership in the local Administrators group for that computer.
  
  
• Dismount the database.
  
  
• Make a copy of the database files that you want to repair.
  
  

Before running Eseutil repair mode on a transport server queue database on a Hub or Edge Transport server role, do the following:

• Stop the Microsoft Exchange Transport service.
  
  
• Make a backup of the queue folder on your Exchange server.
  
  
• Check that you have full permissions for the folders you want to perform the repair on.
  
  
• Make sure you log on by using an account that is a member of the local Administrators group on that computer.
  
  

  Important:
  To keep mail flow uninterrupted, the existing queue folder can be renamed and a new folder and queue database are created when the Microsoft Exchange Transport service is restarted. For more information about doing this, see Working with the Queue Database on Transport Servers.

To run Eseutil /P on a mailbox, public folder, or transport server queue database

Remember the following points after you have run Eseutil /P to repair your mailbox or public folder database:

• Perform a full backup of the database as soon as possible after a repair. Repair invalidates previous backups. This does not mean that previous backups cannot be restored or are without value. It means that the repair makes it impossible to completely roll the database forward from a previous backup. If you restore a previous backup, transaction log file replay will end at the point at which the repair was done. Any changes to the database subsequent to the repair cannot be put back into a restored database. Therefore, it is critical that you perform a full backup of the database as soon as possible after a repair.
  
  
• Remember that you must run Eseutil defragmentation mode (Eseutil /D) and the Information Store Integrity Checker (Isinteg.exe) with the -fix option to finish the repair. Only if you intend to use the repaired database for salvage and then discard it can you skip these extra steps. Skipping these steps means that you may salvage less data than if you completed them, but it can also mean saving several hours of recovery time.
  
  

  Important:
  Before running a repair, you should make a copy of the database by backing up the database files, in case the repaired database is in a worse condition compared to the backed-up version.

You can see the complete command-line reference and syntax for Eseutil /P by typing eseutil /? at a command prompt and then by selecting P for repair.

Note:
In the command-line Help for Eseutil /P, you may notice that switches applicable to streaming (.stm) files, the /createstm switch and the /s switch, are displayed. However, the switches do not work for Microsoft Exchange Server 2007 databases because streaming files are not supported by the Eseutil tool.

For more information about Eseutil, see the following topics:

• Eseutil /D Defragmentation Mode
  
  
• Eseutil /C Restore Mode
  
  
• Eseutil /R Recovery Mode
  
  
• Eseutil /G Integrity Mode
  
  
• Eseutil /M File Dump Mode
  
  
• Eseutil /K Checksum Mode
  
  
• Eseutil /Y Copy File Mode
  
  
• Reference for Common Eseutil Errors