The recovery log helps to ensure that a failure (such as a system power outage or application error) does not leave the database in an inconsistent state. The recovery log is essential when you restart the Tivoli® Storage Manager or the database, and is required if you must restore the database.
When you issue a command to make changes, the changes are committed to the database to complete. A committed change is permanent and cannot be rolled back. If a failure occurs, the changes that were made but not committed are rolled back. Then all committed transactions, which might not have been physically written to disk, are reapplied and committed again.
During the installation process, you specify the directory location, the size of the active log, and the location of the archive logs. You can also specify the directory location of a log mirror if you want the additional protection of mirroring the active log. The amount of space for the archive logs is not limited, which improves the capacity of the server for concurrent operations compared to previous versions.
The space that you designate for the recovery log is managed automatically by the database manager program. Space is used as needed, up to the capacity of the defined log directories. You do not need to create and format volumes for the recovery log.
Ensure that the recovery log has enough space. Monitor the space usage for the recovery log to prevent problems.