Article ID: 258529 - View products that this article applies to.
This article was previously published under Q258529
When you attempt to start the Exchange information store service, you may receive the following event in the Event Viewer application log:
Event Type: Error
Event Source: MSExchangeIS
Event Category: General
Event ID: 9519
Time: 3:30:48 PM
Error 0xfffffde0 starting database /O=Organization/OU=AdministrativeGroup/cn=Configuration/cn=Servers/cn=SERVERNAME/cn=Microsoft Private MDB on the Microsoft Exchange Information Store.
Event Type: Error
Event Source: NTBackup
Event Category: None
Event ID: 8012
Time: 3:12:58 PM
The 'Microsoft Information Store' returned 'The restore environment information isn't found.
' from a call to 'HrESERestoreReopen()' additional data 'SERVERNAME'
In the first event, Error 0xfffffde0 refers to JET_errSoftRecoveryOnBackupDatabase. This means that the Storage Group attempted to restart after a crash, and that the database was restored (but not hard-recovered). In other words, the hard recovery log file replay process was not successfully completed, or was possibly interrupted.
The second event description indicates that the proper Restore.env file could not be located.
This issue can occur if you restore two Storage Groups in the same backup session, and enter the same log file replay location.
Only restore one Storage Group at a time, or if you are restoring multiple Storage Groups, be sure to enter unique paths for the temporary location for log and patch files.
For example, using Microsoft Windows 2000 Backup utility, Ntbackup.exe, you restore a Storage Group (First Storage Group), and enter C:\Temp as the folder for the temporary location for log and patch files, but do not click Last Restore Log Set. This restores the databases to their original location, the log files to the C:\Temp folder, and writes the Restore.env file to the C:\Temp folder as well. Because you did not click Last Restore Log Set, log file replay (hard recovery) will not begin automatically.
If you restore a different Storage Group (Second Storage Group), and enter C:\Temp as the folder for the temporary location for log and patch files, and you then click Last Restore Log Set, this restores the databases to their original location, but overwrites the previous restore session's log files and the Restore.env file.
When hard recovery begins on both Storage Groups (when the information store service is started), the hard recovery process will fail for the First Storage Group, because its log files and Restore.env file are overwritten by the restore process of the Second Storage Group.
Article ID: 258529 - Last Review: December 3, 2007 - Revision: 3.5