Restoring the RASD

If the RASD becomes corrupt (for example, because of a device failure), you can restore the database from the most recent backup files.

The Replication Agent retrieves the RASD backup files from the directory identified by the rasd_backup_dir configuration parameter. See the Replication Agent Reference Manual for more information about the rasd_backup_dir parameter.

1. Log in to the Replication Agent instance with the administrator login.

2. Use the following command to determine the state of the Replication Agent instance:

   ra_status
   

3. If the Replication Agent is in Admin state, skip this step and go to step 4.

   If the Replication Agent is in Replicating state:

   1. Use the following command to suspend replication by the Replication Agent instance:

      suspend
      

   2. Use the following command to verify that the Replication Agent is in Admin state:

      ra_status
      

4. After you verify that the Replication Agent is in Admin state, restore the RASD:

   rasd_restore backup_directory
   

   Here, backup_directory contains the backup you want to restore (for example, 2009-07-04_14.21.34). If you issue rasd_restore without any parameters, only the most recent backup is restored.

   After the restore operation completes successfully, the Replication Agent returns a message to confirm that the RASD restore was successful.

   If the Replication Agent cannot find the directory identified in the rasd_backup_dir parameter, or if it cannot read the RASD backup files in that directory (for example, because of a permission problem), it returns an error. You must correct the cause of the error to restore the RASD.

5. After the RASD is successfully restored from the most recent backup, use the following command to resume replication in the Replication Agent instance:

   resume