4 4 3 Beefalo:Restore
The SEP sesam restore is a process that browses for different versions of savesets and copies data from savesets to a client. SEP sesam provides two restore interfaces: the GUI restore wizard and the web-based Restore Assistant.
This article describes the features and functions of the GUI restore wizard. It is also possible to restore savesets using the Restore Assistant web interface, if you have the appropriate permissions. For details, see Restore Assistant.
The restore procedure is configured using a restore task, where savesets are selected by date/time and task name/file name, and the restore type and restore target (e.g., a file system folder) are specified. Restore granularity depends on the selected data type. Restore tasks can be saved, scheduled and run automatically at regular intervals to check and control data integrity. For details, see SEPuler.
Restore in GUI is triggered from the Menu bar -> Activities -> Restore. The restore wizard guides you through the process of restoring your data. For step-by-step instructions, see Standard Restore Procedure.
To learn how to recover server data from savesets after data loss due to a disaster, see SEP sesam Server Disaster Recovery.
The following restore types are available:
- Complete restore
- A process by which an entire saveset is restored – all the files of a saveset are selected and restored.
- Selective restore
- Only a subset of a saveset's files is selected – only selected parts of data on the saveset, such as individual files and directories, are restored.
- Generation restore
- All generations of backed-up files, including differential or incremental backup files, can be browsed for and selected for restore. SEP sesam automatically switches to the generation view in which all the generations of the files are presented. The selection list displays all files since the last full backup.
Scheduled restores are performed by creating an event or task. It is also possible to schedule a restore for data that has not yet been backed up. The selective restore can be set to run at the completion of the daily backup routine, e.g., the last backed up file can be restored, preferably to a new directory and the data validity established. For details, see Scheduling Restore.
The restore wizard can be accessed from the menu bar -> Activities -> Restore. The following restore options are available in the restore wizard to define how your data will be restored.
Step 1: Welcome page
By default, the welcome page is not displayed when the wizard starts. It contains only general information on specifying the restore process.
Step 2: Filter
You can select how you want to search for your backed up data. Savesets can be searched by the task name in the saveset or by filename or path.
You can select among tasks for which backups are available from the Task drop-down list or enter a regular expression in the Search pattern field to search for files by their name or path. Note that the search may take some time (depending on the number of processed tasks).
Saved in period allows searching for savesets only within the specified time frame. Show external results extends the search to the savesets generated with supplementary backup tools and interfaces, such as BSR Pro, MySQL, DB2, etc.
Step 3: Select task
Select a task for the restore.
The following options depend on the type of data (task type) you want to restore. You will see only see the options that are relevant to your data restore.
- Complete restore of selected task will restore everything in the task.
- Hide migrated means that the savesets that have been moved with a migration task will not appear in the results.
- Normally, Generation restore is selected by default. Make sure this option is enabled if you are restoring from an incremental or differential backup.
- As path restore (corresponds to the Restore Assistant option Restore (any) backup to file system) should be enabled if you have backed up data using a task type other than default Path, such as Hyper-V, and you want to perform a simple path restore of your VM data directly to a directory on any system.
- Disaster restore should be enabled if you want to restore a server after a disaster scenario.
|As of v. 5.0.0 Jaglion, the Hide migrated option is no longer available. The migrated savesets are now displayed only if there is no original (backup) saveset. If you want to use the migrated saveset instead of the original saveset, you need to select it as Preferred Media pool in the Target Settings window (step 5).|
For details on disaster recovery, see SEP sesam Server Disaster Recovery. If a client server is being recovered, simply select the server’s disaster recovery backup task.
Option Mount saveset – Cross-Platform Recovery File System Layer (XPRFS)
The checkbox Mount saveset under the Single File Restore via Mount is available after selecting the saveset for restore. You can mount any non-compressed and non-encrypted file-based saveset stored on SEP sesam data store (Path, Si3, etc.). Mounting is not possible if the savesets are stored on tapes. In v. ≥ 5.0.0 Jaglion, you can also select RDS to mount on from the drop-down list and specify advanced mount options. For more details on mount requirements and limitations, see Cross-Platform Recovery File System Layer.
Step 4: Select files
You can select individual items for restore.
The Search button below provides an additional option to search for files within the target saveset.
Step 5: Target settings
This step provides an overview of the restore target settings. These settings differ slightly when restoring a virtual machine. The saveset selected for restore is displayed with its attributes at the top. In the Restore task name field, the name of the task is assigned automatically and can be modified.
Most of the following options already have default values or are enabled by default and can be used in most cases. Some allow fine-tuning, which is required for more specific data restore.
Target node: The name of the server where the files will be restored.
The Target path settings provide the following options, which differ slightly for VM restore:
- Restore to original target path or Restore to original VM (for VM restore) is selected by default and restores the data to the same location from which it was backed up. Note that this only applies if the target node is the same server from which the data was backed up, i.e., if the source and target server are the same.
- The Original tree structure checkbox is selected by default when restoring to the original location. The directory structure of the restored files is the same as the original directory structure of the backed up data. If this option is disabled, the directory structure is not recreated and only the files are restored (flat structure).
- Select New restore target or for VM Restore with new VM name (formerly Change VM target selection) if you want to restore the data to a different location. You can manually enter the path to the directory where the data is to be restored or use the Browse button.
- If you choose Restore with new VM name, use the Browser button to select the restore location for your VM. Expand the Change VM target selection option to set additional options, e.g., define the target node or set data store path.
Execution options or VM restore options (for VM restore) provides the following options:
- Do not overwrite existing items is selected by default. Files are restored only if they do not already exist on the target system.
- Overwrite existing items replaces the files on the target system with their restored version.
- Create a new version restores the files under a new name (not available for all restore types).
When performing a VM restore, you can also specify whether to start a virtual machine after restore.
Optional data source selection:
- Preferred media pool: If the data was migrated or replicated and different media pools were used for the process, the source media pool to restore from can be selected from the drop-down list of available media pools.
- Used media | barcode: If the data was migrated, the migrated copy is stored to another media. Here you can select the source media from which to restore from the drop-down list of available media.
- Drive: The number of the drive with the backed up data (more relevant for tape backups without a loader). Optionally, select the drive number of the drive to be used for reading the data. Typically, you use this if you have configured additional drives and want to assign dedicated drives for restore. For details, see the option Create second drive in Configuring a Data Store.
- Interface: The name of the server from which the files are being restored. This is either a Remote Device Server (RDS) or the SEP sesam Server.
When you click the Expert options, a new window opens with different tabs where you can set more specific options and actions.
|The Expert Options button for setting advanced restore options is available only in advanced mode (formerly expert GUI mode). If you are running the GUI in simple mode (formerly basic GUI mode) and want to access Expert Options, change simple UI mode to advanced, as described in Selecting UI mode.|
The Verbose level allows you to change the verbosity of the main and trace logs to display more information than in standard logging mode. For details, see Setting Log Level.
|Setting||Log level details|
|0||print only standard and error messages together with a summary|
|1||display a line for every item when its processing starts: sbc-3008: Info: Processing item: [xxx]...|
|2||display a line for every item when its processing finishes: sbc-3108: Info: Item processed successfully: [xxx]|
|3||display backup module processing information (with DB_API modules)|
|4||display underlying module processing XBSA and detailed DB_API modules|
|5||display packing data (mtf, cpio, sidf) trace messages|
Restore options text box allows you to enter other specific options.
Based on the specification in the previous steps of the wizard, one of the following is already selected: Restore the complete saveset, Restore according to the composed list or disk_info disaster recovery. Changes of these restore options override the specifications defined in the previous steps.
- Restore the complete saveset restores all files in the saveset.
- Restore according to the composed list performs a selective restore according to regular expression-based filtering; see Filter tab below.
- disk_info disaster recovery restores disk information, e.g., a master boot record and the registry during a disaster recovery.
By default, No relocation is selected; this restores savesets to the original target path. If the With relocation option is enabled, you have to specify a new target location for the restored files. Individual files can be restored to different paths.
Files for selective restore can be defined using pattern-based filtering.
Task type tab
By default, the task type by which the data is backed up is selected (default is Path; other task types include Hyper-V, SAP Hana, Exchange Server, Lotus Notes, MySQL, VMware vSphere, etc.).
- If you have backed up data with task type other than the default Path, e.g., Hyper-V, you can still use the option As path backup. This option allows you to restore your data directly to the file system without any additional action.
- The option Write saveset into file (corresponds to the Restore Assistant option Write backup into dump file) allows you to restore the entire saveset to a single file rather than to its original location.
|To use a pre/post interface, you must first activate it and create your own commands. Interfaces can be activated in GUI: in the menu bar, click Configuration -> Interfaces -> select the interface you want to activate and the client. The interface script window opens, allowing you to configure the desired action. Once done, click Save to activate the interface. The interface is saved to the <SESAM_ROOT>/bin/sesam directory. For more details, see Using Pre and Post Scripts.|
Retention Time tab (formerly EOL)
Specifies for how long (in days) this restore task is kept and can be accessed by using GUI. The default value is 7 days. The value 0 means that a restore task is kept permanently unless you manually delete it.
Execution Time tab (formerly Start time)
- Instead of starting your restore immediately, you can specify the start time by entering the desired date and time.
- Start time frame: You can specify the maximum allowed start time (the amount of time an event can be in the queue) in days and/or hours. The actual start time of the event depends on the processor load of the system. At the start time of the event, every event is forwarded into the queue, where it might wait until jobs with a higher priority finish. For jobs with the same priority, the FIFO rule (first in first out) applies. If the event cannot be executed within this specified time frame, it will not be executed at all.
- Cancel execution after (formerly Stop task if it runs longer than): You can specify the amount of time (days and/or hours) after the event's scheduled start time, after which the event will be cancelled.
- Cancel execution at (formerly Stop task at): Depending on the previous setting, it specifies the exact date and time when the task will be stopped.
- Follow up: You can specify the follow-up event to trigger the predefined event on the SEP sesam Server once the restore is completed For details, see Follow-up Events.
Step 6: Summary
In the final step of the restore wizard, review the summary of your restore configuration. Click Start to run the restore or Save to save the task under the assigned name in the Name field.
|All, started and scheduled restores are saved in SEP sesam under Main selection -> Tasks -> Tasks by clients/Tasks by groups with the current timestamp. A restore can be restarted via the menu item Activities -> Restore or via Main selection -> Job state -> Restore -> right-click selected restore -> Immediate start.|