立即与支持人员聊天
与支持团队交流

LiteSpeed for SQL Server 8.9 - Configure Log Shipping Guide

Restore Databases Using the Restore Wizard

The Restore wizard guides you through the process of restoring a database (full or differential), transaction log, files, or filegroups.

Tip: Database backups created with LiteSpeed Version 8.x cannot be restored using older versions of LiteSpeed.

NOTES:

Scenario

You need to restore a LiteSpeed disk backup to a new database on another SQL Server. Copy the backup files needed for restore to another server and run the Restore wizard. ClosedClick here to see additional information for this scenario.

To run the Restore wizard

  1. Select the Backup Manager pane (CTRL+1).
  2. In the Navigation pane, select a database to restore and click the Restore or Automated Restore button. Alternately you can right-click a database and select Restore or Automated Restore from the menu. If you clicked Restore, click one of the following. If you clicked Automated Restore, review the automated restore information below.

    • Database (full or differential)—Restore database only.
    • Files and Filegroups—Restore files and filegroups only.
    • Transaction Log—Restore transaction log only.

      Tip: The database (full or differential), files and filegroups, and transaction log restore types use the restore wizard pages listed in the steps below: restore destination, backup source, backup content, recovery options, data files, scheduled restore, execute script, and finish.

    • Attached Files—Restore attached files only.

      Tip: The attached files type uses the restore wizard pages listed in the steps below: restore destination, backup source, backup content, data files, attached files, scheduled restore, execute script, and finish.

    • Automated Restore—Restore the most recent full backup and optionally differential and transaction log backups. If you select this option, the Restore wizard creates an Automated Restore job. Backup files can be restored immediately or restored in the background.

      Tip: The automated restore type uses the restore wizard pages listed in the steps below: restore destination, backup source, backup files, restore options, database integrity, database files, scheduled restore, notification, execute script, and finish.

  1. Review information on the Restore Destination page. Select the Kill all current connections before restore checkbox to obtain exclusive access to the selected database. Additionally, you can select the View current activity link to view the database connection activity table. A database cannot be restored unless the restore process has exclusive access to the database. No user connections can exist when performing a database restore.

    Caution: Automated Restore for multiple databases - When running restores for multiple databases, we recommend that you use the %DATABASENAME% variable to automatically generate an original database name for database(s) going to be restored. When running restores from multiple servers, different databases with the same server name can be overwritten. To prevent this, the variable %SOURCESERVER% adds the source server name (server+instance) to the target database name when running multiple automated restores. For example: %SOURCESERVER%_%DATABASENAME%.

  2. On the Backup Source page, select Database to restore from a specific database's backup history, or select Device to manually select files to restore.

    Database

    If you select Database, the database you are restoring is displayed. You can select another database using the Database drop down selector. Use the SQL Server drop down selector if the database is on a different SQL Server instance.

    TIP:You can do point in time restores, restoring from a specific date and time or a time period (for example four hours prior to recovery).

    TIP: The restore type option Restore Verify Only in combination with the backup type Verify the latest full or differential backup is less time consuming for full and differential backups or fast compression backups where you only want to verify the latest backup, rather than the full and the latest differential. For backups that are run every day this will serve to verify all backups.

    Device

    If you select Device, the options are Disk, Cloud, TSM Backup, TSM Archive, and Tape.

    If you selected Cloud, review the Cloud Account Settings information below:

    Cloud account

    Select the Cloud account from the drop-down list.

    Select the following items to add or edit the registered cloud account settings.

    • Add - (For Amazon S3 only), click to add cloud vendor, display name, authentication, region, storage class (standard, infrequent access, reduced redundancy storage), and bucket. Select: use SSL, use server side encryption (AES-256), GovCloud (US) Region, and automatic striping (auto, 10, 25, 50, 100, 500, 1000, 1995) GB. Select 'Use Amazon S3 Transfer Acceleration Speed' to use Amazon's S3 Transfer Acceleration feature which allows for increased upload speed to S3 storage up to 200% in some cases by using local CloudFront endpoints.
    • Add - (For Azure Blob only), click to add cloud vendor, display name, storage account name, access key, storage type (block blobs or page blobs), container, use SSL, government account, and automatic striping . Options for block blobs are: auto, 10, 25, 50, 100, 250, 500, 1000, 2000, 3000, 4000 and 4300 GB. Options for page blobs are: auto, 10, 25, 50, 100, 250, 500, and 995 GB.
    • Add - ( For Google Storage only), click to add cloud vendor, display name, service account ID, private key, project ID, storage class, region and bucket. Use SSL is always selected as Google always uses it.
    • Edit - (For Amazon S3 only), click to edit display name, authentication, region, storage class (standard, infrequent access, reduced redundancy storage), and bucket. Bucket name must conform to DNS naming requirements and must not contain periods ("."). Select: use SSL, use server side encryption (AES-256), GovCloud (US) Region, and automatic striping. Options for automatic striping are: auto, 10, 25, 50, 100, 250, 500, 1000, and 1995 GB.

      Select 'Use Amazon S3 Transfer Acceleration' to use Amazon's S3 Transfer Acceleration feature which allows for increased upload speed to S3 storage up to 200% in some cases by using local CloudFront endpoints. Additional data transfer charges may apply. See Amazon S3 pricing for more details.

    • Edit - ( For Azure Blob only), click to edit display name, access key, storage type (block blobs or page blobs), container, use SSL, government account, and automatic striping. Options for block blobs are: auto, 10, 25, 50, 100, 250, 500, 1000, 2000, 3000, 4000 and 4300 GB. Options for page blobs are: auto, 10, 25, 50, 100, 250, 500, and 995 GB.
    • Edit - ( For Google Storage only), click to edit display name, service account ID, private key, project ID, storage class, region and bucket. Use SSL is always selected as Google always uses it.
    • Delete - Click to delete the Cloud account from the console.
    • Import - Click to import a saved Cloud account in XML format.
    • Export - Click to export and save a Cloud account in XML format.

    Proxy Settings

    Select the following items to edit the Cloud account proxy settings.

    • Use LiteSpeed Server proxy settings - Click to use the server proxy setting. This is the default selection. You can also edit the proxy settings from this item.
    • Use LiteSpeed Console proxy settings - Click to use the console proxy setting. You can also edit the proxy settings from this item.
    • Specify custom proxy settings - Click to add your own custom proxy address, port, username and password.

    If you selected TSM Backup or TSM Archive, review the following for additional information:Closedclick here for field descriptions.

    If you selected Automated Restore, review the following for additional information.

    Restore from

    Select SQL Server and databases to specify the source from where to search the backups for automated restore.

    NOTE: The restore from parameters help LiteSpeed to narrow down its search for the required backup files in the source folders.

    Use the drop-down treeview to add multiple sources from the System and User parent nodes. Select User databases and all subordinate user databases are automatically selected. Individually select and deselect databases using the checkbox next to it.

    For AlwaysOn availability groups, you may need to specify both SQL Server (primary and secondary) to allow LiteSpeed to search backups among all replicas. To specify a secondary SQL Server, set "Automatically add all the Availability Group replicas to the search list" option.

    Latest backup search method

    Select one of the following options:

    • Folder scan—LiteSpeed will search the specified folder for the most recent database backups. You can configure LiteSpeed to search subfolders and filter backups using the specified file extensions.

    • Specific backup file name—You can use this option to automate restore operations if the same file gets overwritten during the backup or if new backups are always appended to the same file.

    NOTES:

    • If restoring a striped backup, you can specify multiple locations/filenames.
    • You can enter several backup extensions per path. Separate them with commas or semicolons.
    Backup type

    Specify backup types to use for the restore. Select one of the following options:

    • Full—The most recent full database backup.
    • Full and differential—The most recent full database backup and any existing differential backups based on this full.
    • Full, differential and transaction logs—The most recent full database backup and any existing differential and/or transaction log backups created after the most recent full backup.
    • Include copy only backups—Select the Include copy only backups check box to add copy only backups in the restore.
    • Specify a point in time to restore to. By default LiteSpeed restores to the most recent state possible. Alternatively, specify a specific date and time or a date and time relative to the restore time. For example, specify a time measured in days, hours, minutes and seconds from the restore time.
  3. If you selected Automated Restore, on the Backup Files page, select the backup file locations to search.

    Restore from

    Select the restore from location (disk or cloud) using the drop down menu.

    Add

    Add the backup file location to restore to the list by selecting and using the Backup File Location wizard.

    Remove

    Select and remove the backup file location from the list.

  4. If you selected Automated Restore, on the Database Integrity page, define the options to check database integrity after restore.

    Check database integrity after restore (DBCC_CHECKDB)

    Use this option to run a CHECKDB on the restored database and report the results to the repository for review. This option is selected by default. Select a combination of the following database integrity options:

    • Check physical structure only (PHYSICAL_ONLY).This option is selected by default.
    • Check the database for column values that are not valid or out-of-range (DATA_PURITY).
    • Perform logical consistency checks on indexed views. XML indexes and spatial indexes (EXTENDED_LOGICAL_CHECKS).
    • Do not perform intensive checks of nonclustered indexes for user tables. This option is selected by default.
    • Use locks instead of using an internal database snapshot.
    • Do not include informational messages in notification report (NO_INFOMSGS). This option is selected by default.
  5. If you selected Automated Restore, on the Restore Options page, specify options for automated restore.

    Drop databases after restore

    Use this option if you no longer need the restored database. For example, if you are only restoring the latest backup for testing purposes. This option contains two additional options to select. One or both options can be selected.

    • On success restore and check database integrity operations - The database is dropped after a successful restore and database integrity check.
    • On failure any of restore or check databases integrity operations - The database is dropped after failing the restore or database integrity check.
    Overwrite the existing database Use this option if you want to overwrite the existing database with the restored database.
    Include databases that are part of a replication plan

    Use this option to include databases that are part of a replication plan.

    Recovery state

    The options are as follows:

    • Select to leave the database in an operational state (RESTORE WITH RECOVERY). The default is selected.
    • Select to leave the database in a non-operational state and allow restoration of additional transaction logs (RESTORE WITH NORECOVERY). The default is not selected.
    Password

    Provide a password for encrypted backups.

    NOTE: Automated Restore requires that you use the same password for all encrypted backups.

  6. Review the following additional information about the Backup Content page. Skip this page for Automated Restore.

    Select server instance

    Select the server that contains the backup you want to restore.

    Note: In the case of AlwaysOn Availability Group backups, you may select "Any instance" item from the list. It will include backups created on all replicas.

    Select database backup Select the database that you want to restore.
    First backup to recover
    1. Click the ellipsis to launch a window containing a list of backups to restore.
    1. Scroll down the list and select a backup. The backup to recover table is populated with a list of backups. The table includes backup name, type, destination, encrypted, server, database, position, begin date, finish date, size, user, expiration, and copy only. See the partial table below.
    • indicates a successful full backup.
    • indicates a successful differential backup.
    • indicates a successful transaction log backup.
    • indicates a warning. Possible reasons: encrypted backup that requires entering an encryption key, backup no longer exists, or corrupted backup.
    • Red text indicates the backup is not available. This could be because the backup was local to the source database and not available from the target or because the backup does not exist.
    • indicates a selected backup.

    IntelliRestore

    Select to have LiteSpeed automatically select the backups needed to restore the database successfully.

    Verify backups

    Select to verify the backup file integrity before completing the wizard.

  7. On the Recovery Options page, select the database recovery state following the restore. Skip this page for Automated Restore.

    Overwrite the existing database Select to have the current database overwritten with the restored database. The default is not selected.
    Preserve the replication settings Select to preserve the replication settings for the restored database. The default is not selected.
    Restrict access to the restored database

    Select to restrict access to the database after it is restored. The default is not selected.

    Recovery state

    The options are as follows:

    • Select to leave the database in an operational state (RESTORE WITH RECOVERY). The default is selected.
    • Select to leave the database in a non-operational state and allow restoration of additional transaction logs (RESTORE WITH NORECOVERY). The default is not selected.
    • Select to leave the database in read-only mode. Undo committed transactions and save the undo actions in a standby file (RECOVER WITH STANDBY). The default is not selected. When selected, the default listed standby file can be used. Or select the ellipsis button to browse and select another standby file.
  8. Review the following additional information about the Data Files page:

    Prompt before restoring each backup

    Select this option if you would like to receive a prompt notification before restoring each backup. The default is not selected.

    Note: Not available in case of Automated Restore

    Eject tapes (if any) after restoring each backup

    Select this option if you would like to eject any tapes after restoring each backup. The default is not selected. This option is available only for tapes.

    Note: Not available in case of Automated Restore

    Restore as compressed, read-only database

    Using this option, you can restore a user database into an NTFS compressed folder or restore a tlog to a read-only database in a compressed folder.

    NOTES:

    • When using an NTFS-compressed folder for a database, it can only be restored as read-only.
    • You can only use this feature on Windows NTFS file systems.

    Specify a compressed folder for the data files by editing the Restore As paths. If a folder does not exist, LiteSpeed will create it as NTFS compressed.

    Restore the database files as

    Although you can manually enter DATA and LOG locations, including secondary data files locations, it is recommended that you use locations generated by LiteSpeed.

    If restoring database (full or differential), the following links can be selected (not available for Automated Restore):

    • Keep original database (full or differential)—Click to display in the table below the original database locations.
    • Use SQL Server instance default locations—Click to display in the table below the SQL Server instance default locations.
    • Select new location—Click to launch the Database Files Destination window and select a new database output file location. You can browse the network, add, delete, and rename files.
    • Restore to locations from backup set—Click to get locations from the backup set.

    If the source and target locations do not match or if they are set to other than the default, LiteSpeed you can select one of the following options (not available for Automated Restore):

    • Use SQL Server instance default locations—To use DATA and LOG directories of the existing database you are restoring the backup to. The default is selected.
    • Custom locations for data and log files—To use DATA and LOG directories of the database which backup you are restoring. The default is not selected.
    • Folder for data files—To enter a new location for all DATA files. Click the ellipsis button to browse and select other folders for data files.
    • Folders for log files—To enter a new location for all LOG files. Click the ellipsis button to browse and select other folders for log files.

    If Automated Restore selected, the following options can be selected:

    • Use SQL Server instance default locations - default SQL Server values will be used for DATA and LOG files. Check the SQL Server options.
    • Restore to locations from backup set - Automated Restore will use locations from the backup set. The locations must be available at the selected destination SQL Server instance.
    • Custom locations for data and log files — allows to set default folders for DATA and LOG files and optionally set path to all or some DATA and LOG fils by its logical file names. The following fields have to be defined for every optional file path:
      • Logical Name - logical file name of the DATA or LOG file
      • Database - restoring database name
      • Restore As - full path to a new DATA or LOG file location

     

    Processor affinity

    Click to select which processors LiteSpeed can use. The default is 0, which allows LiteSpeed to use all available system processors.

    Logging level

    Select one of the following options:

    • None—LiteSpeed does not write a log file for the backup or restore operation.

    • Verbose—LiteSpeed writes a log file for the backup or restore operation.

    • Verbose. Log file is removed on success—LiteSpeed only saves log files if the backup or restore operation fails. If it succeeds, LiteSpeed does not save the log.

    Network resilience

    If LiteSpeed fails to write disk backups or reads from disk, it waits and retries the operation. You can enable and disable and control the number of times to retry and the amount of time to wait before retrying.

    • Number of times to retry any given read/write attempt—The default is 4 retries. The maximum allowed setting is 1000 retries.
    • Wait period before each retry attempt (in seconds)—The default period to wait before retry is 15 seconds The maximum allowed setting is 300 seconds.

    For more information, see Network Resilience.

  9. If you added an attachment to the backup file, select the Restore Attached Files and Directories on the Attached Files page.

  10. Review the following information about the Schedule Restore page. Select Weekly on (for Automated Restore only), Run immediately, Run in background, or Schedule (Custom Schedule for Automated Restore. Selecting Schedule launches a page for adding the schedule name, schedule type, occurs, weekly, daily frequency, duration, and description.
  11. Review the following information about the Notification page (for Automated Restore only). You can specify the notification of failure options that are sent after each restore. Select one of the following:
    • Do not use notification—All failure notifications (including operator selections) are disabled.
    • Notify every time—Notify for all successes and failures. An operator can be selected and configured to receive notifications. Click the ellipsis button on the far right to create new or edit existing operators.
    • Failure only—Notify for failure only. An operator can be selected and configured to receive notifications. Click the ellipsis button on the far right to create new or edit existing operators.

      NOTE: The SQL Server Agent must be configured to send email using Database Mail. Review the following for additional information:

  12. Complete the wizard.

 

Related Topics

相关文档

The document was helpful.

选择评级

I easily found the information I needed.

选择评级