Details

Tape Migration Frequently Asked Questions

What is the difference between tape migration, twinning, and duplicate?

The ability to migrate data from one form of media to another was introduced in DPX. This feature allows you to migrate data from the original target location to another target. It can be the same type of media or any other media available to the backup environment. This feature differs from the twinning and duplicate facilities because the migrated data has its own expiration time apart from the original backup. In the case of twinning, which creates two copies of the backup simultaneously, the expiration for all media is the same and cannot be changed. Similarly, in the case of the duplicate facility, the only difference from twinning is that the media is copied from the original media to like media at a separately scheduled time or immediately.
 

Are there any prerequisites to tape migration?

The master and all device servers involved in the migration should be at the latest patch level. It is imperative that at a minimum they are at the same patch level since there is a great dependency between the master and device servers for a successful migration. Also, a device server cannot be a NetApp storage system attached tape library or tape device.
 

Where is the tape migration job configured?

A tape migration job is configured on the Manage Tapes tab.
 

 A Tape Migration can also be included as part of the schedule when configuring a job. This is only available for File, Image and NDMP backups.
 

 

 

What options are available for migrations on the Manage Tapes tab?

Tape migration allows you to migrate by job or by media. The view is changed by right-clicking the Enterprise. A pop-up menu appears and you can toggle the view by job or media. When selecting migration by media view, the following window appears:
 

 

 
When you expand the Media Pool icon, all the media pools in the environment appear. Expanding the media pools displays all of the volsers within each pool with a FULL or APPENDABLE status. You can also expand each volser to see which jobs are currently available. You can only select the entire volser. To migrate a job, the view must be by job and not by media. More than one volser can be selected, but they must be present in the same device cluster since only one device cluster can be selected per migration job. The remainder of the job setup is the same as any backup job going to media. It can be run at a scheduled time or immediately.

When selecting to migrate by Job, the administrator consolidates a selection of backups for a specific job to a single tape or set of tapes. The limitation is that the job can only be for a specific job; multiple jobs cannot be selected when configuring the job. The following image shows what displays in the management console when the view is toggled to job.

 
The following image shows what displays in the management console when the Backup Jobs icon is expanded and the jobs themselves are expanded. All available backup instances for the expanded job display. Only backups under one job can be selected for migration. If more than one job needs to be migrated, then a separate migration job must be created for each. The remainder of the job setup is the same as any backup job going to media.
 

 

 

How do I schedule a migration job as part of the backup job?

A migration job is scheduled as part of a backup job as well. Selecting Schedule Backup Job presents the pop-up window shown in the following image. Expanding the Run drop-down menu shows the Migrate option.

 

 
Once you select Migrate, the Job Schedule window appears as follows. Clicking the drop-down next to Migrate presents two options: to run when the backup job completes, or to run as per a schedule.

 
Before applying the schedule, you must select a Destination for Migration. There are two drop-down menus, one for the Device Cluster and another for the Media pool. As with any backup job to media, the Device Cluster and Media pool must be the same type.
 

 

 
The Migrate Options window displays information similar to the Set Destination Options for the job itself. You can set a retention period as well as other options for the migrated information, or keep the default settings.
 

 

 
After selecting Upon Job Completion, filling in the Destination for Migration information and clicking Apply, the schedule appears as follows. At this point, click OK and save the job. When the backup job runs, the migration starts when the backup completes.
 

 

 
If the migration is to be scheduled, then the image following shows what is seen in the pop-up window. The scheduling of this type of migration is the same as a standard backup job to media. The following example uses the Daily template, but any available templates can be used. It is possible to schedule multiple migrations using the schedule feature but it is not possible to send them to different Device Clusters and Media pools. The first job scheduled takes precedence and all others use the same Device Cluster and Media pool.

Why would I use the migration capabilities of DPX?

The benefits of the Migration facility with DPX depend on the existing requirements within an environment. One example is the retiring of older forms of media to accommodate improved or upgraded technology. The need to maintain data that is already protected on older storage platforms can be challenging as the older media can be difficult to maintain. Migration allows the preservation of this data by not only moving the data itself from the legacy platform to a newer platform, but maintaining the information in the Catalog for easy retrieval.
 
Another example is a backup that uses more than a single tape device simultaneously, resulting in small amounts of data spread over several media locations. A single job may be present on multiple tapes to improve the backup window performance. A migration can consolidate the information to a smaller number of backup devices.  
 
The need to offsite media that has a longer expiration time then what is maintained locally is another example of the benefits of the migration facility. The target media can be given its own retention period apart from the original job.
 

What do I do if my migration fails?

The most common cause for a failed migration is the media that is being migrated is not present on the device server. Other causes include insufficient resources to perform the migration or running out of target media. If you cannot determine the reason for the failure, Catalogic Technical Support can assist in determining what is needed to resolve the issue. There are times when a migration may fail and it is determined that the data no longer needs to be migrated. The migration that is part of a regular backup job attempts to migrate that information every time it runs. To resolve this situation, remove the scheduled migration task of the job, save the job, then add the scheduled migration back into the job. This resets the migration job and tries to migrate only the data protected by the most recent job, and not data that was not migrated due to a previous issue.