Database Recovery

ESE has two types of database recovery: hard and soft.

Hard Recovery

A hard database recovery is a manual process that occurs when transaction log files must be replayed into a restored online backup using a utility such as ESENTUTL. CygNet Software recommends that you use the DBS Restore Backup utility to perform hard recovery.

The topic Hard Recovery describes how hard recovery works for an ESE-based datastore.

Also see Recovery for information about the Recovery options of the ESENTUTL utility.

Soft Recovery

A soft database recovery is an automatic process that occurs when the service is restarted. The transaction log files in the service’s txlogs folder are replayed. ESE uses the checkpoint file to determine which transaction log file to start with when it sequentially replays transactions into databases. The process reverses any uncompleted transactions (per the atomic nature of the database) and writes any transactions that have not already been written. This process makes the databases up-to-date with all recorded transactions.

A fundamental assumption of the soft recovery process is that no database or log files have been moved, deleted, or destroyed by the failure or the administrator after the failure. If the logs are unavailable, you must manually repair the database by running ESENTUTL to make the database consistent and ready to start. See Repair for instructions on using the ESENTUTL utility.

Recovery Event

Below is an example of what would be recorded in the Event Viewer in the case of a soft recovery.