The space available in Exchange Information Store databases exists a list of pages. This space is called space tree, which is held as binary tree. An Exchange Server starts consuming high CPU percentage and showing slow performance factors if this space tree gets highly fragmented. Such problems can be resolved by running eseutil /d command, which shrinks the size of space tree and thus, provides offline defragmentation. On occasion, the process is left half-way through. If this event occurs, it could prevent Information Store from starting for the reason that database is corrupt. To restore the database, you are suggested to apply your last online or offline backup. If this too fails, you should use Exchange Server Recovery software to scan and repair it.
To exemplify, you might come across an error message 4294966277 after you run offline defragmentation using eseutil /d command. The information store fails to start after you receive this error. When you view the application event log, it reveals error 0xfffffc05 while initializing the database. The event is marked with 1120 ID.
Cause
When you perform offline defragmentation of Exchange database, the database is first copied to a temporary file to perform the procedure. After it completes successfully, the defragmented database is them copied back to the original location to overwrite the original database.
The above error occurs when the tool doesn’t complete the copy back operation of temporary database to original location. This leaves the database in damaged state.
Solution
The temporary database file is not deleted until the copying operation completes. So, you can use it to manually copy the database, provided it is intact.
-
For a single database, you can execute eseutil /g d:\tempdfrg.edb command to check the consistency of the temporary database file. If it gives clean results, you can copy the database to D:\Exchsrvr\Mdbdata folder after renaming the original one. You should then rename the copied database to correct database.
-
If you defragmented both Priv.edb and Pub.edb databases and affected database is unknown, you should first check them for integrity using eseutil /g command. The database that fails the test should then be checked for size and compared with the size of temporary file. You can then copy the affected database by following steps, as above.
If the temporary database fails the integrity check, it is damaged too. The database corrupted in this way should be restored from backup. However, when no backup is available, you need to opt for powerful third-party Exchange Recovery solutions. Exchange Server Recovery software are intelligent utilities to scan and repair damaged Exchange databases, while maintaining their integrity.
Stellar Phoenix Mailbox Exchange Recovery is a feature-rich utility that employs powerful scanning algorithms to repair a damaged Exchange database. It supports Exchange Server 5.5, 2000 and 2003. It is a safe Microsoft Exchange Recovery utility with impressive features. The tool extracts user mailboxes in individual .pst file format files.
