100 percent restored. [SQLSTATE 01000]
Processed 52345768 pages for database ‘SQLDBPool’, file ‘SQLDBPool_Data’ on file 1. [SQLSTATE 01000]
Processed 3045777 pages for database ‘SQLDBPool’, file ‘SQLDBPool_Log’ on file 1. [SQLSTATE 01000]
Many times you came across a situation where the Database restore is completed but you can see the restore command is running for a long time and your database status is restoring. We always want to know why database restore is taking so long time.
Let’s first check out what restore is doing internally. Restore session performs below three tasks while you execute the restore command.
- Data Copy Phase
- Redo phase
- Undo phase
The data copy phase involves copying all the data, log, and index pages from the backup of a database to the database files. After the completion of this phase SQL Server reports restore completes 100 percent.
In the Redo phase, all the committed transactions present in the transaction log when the database was backed up are rolled forward. It means all the committed transaction changes applied to the database. Still the database has un-committed transaction so database will be in unusable state.
In the Undo phase, all the uncommitted transactions in the transaction log while the database was backed up are rolled back. If the database is being restored with NORECOVERY, the Undo phase is skipped.
From the above explanation you can understand why the restore command is taking longer time after 100 percent completes.