The IBM ProtecTIER(R) product offers replication capabilities, allowing administrators to copy data among one or more local and remote sites. Each site has a single or clustered ProtecTIER environment, which contains one or two ProtecTIER nodes and a storage repository. Replication occurs among repositories that are part of the same replication grid.
You can use IBM ProtecTIER Manager software to set policies for replicating data across repositories. And because the ProtecTIER product deduplicates data before storing it, only the changed data is transferred to the remote site.
In this e-Learning module, you will learn about replicating virtual tape library, or VTL, repositories. We will discuss the concept of replication; how to perform key replication tasks, including creating a replication grid, defining repositories, and establishing replication policies; and how to perform disaster recovery operations. For information about OpenStorage and file system replication, see the other e-Learning modules.
Replication occurs among local and remote sites. Data can be replicated between single ProtecTIER nodes, between clusters - with each cluster containing up to two ProtecTIER nodes - or between single nodes and clusters. The replication flows in both directions between repositories.
There are several steps to take before you can perform replication. First, configure the system for replication. Next, add a Replication Manager using the IBM ProtecTIER Manager interface. Then, log in to the Replication Manager, and create a replication grid. Finally, add repositories to that grid, and create a replication group. We will now look at each of these steps in more detail.
After you set the time and date and complete other configuration tasks, use the ProtecTIER Service Menu to configure replication. If you are working with a TS7610 or TS7620 system, this step is required. For other systems, this step was completed by IBM representatives as part of your initial setup process.
To configure replication, use CLI commands to configure the ProtecTIER Replication Manager application on the server that will remotely manage the replication grid. You can also update the replication network with IP addresses and configure the static routes among servers using Replication Manager.
For more details, see the Installation and Setup Guide.
Your next step is to open the IBM ProtecTIER Manager interface to add Replication Manager. Click Grids Management, and click Replication Manager > Add Replication Manager. Enter the IP address for the new Replication Manager.
Next, click Create new grid. Note that if you installed Replication Manager on a ProtecTIER node, you are limited to one grid. The grid ID is automatically generated by the Replication Manager, but you can change it if you prefer. Enter a grid name, and click OK.
To add a repository to the grid you just created, click Repository > Add repository to grid. Enter the replication IP address and port information and the administrator user name and password. This is the same user name and password that you enter when you log in to perform system management tasks.
Click OK, and a stand-alone repository is added, which indicates that it is not yet part of the replication group.
Repeat this step to add your remaining repositories to the grid, so that at least two repositories are displayed.
Now, you can create a replication group. Select Group > Replication group management. The Replication group wizard is displayed. Select one or more repositories to include in the group. Then, preview your selections. The arrows indicate the bidirectional flow of data that occurs during replication.
In the Replication Manager panel, select a repository to view additional information in the right pane, including physical and nominal capacity, replication activity and alerts, backup activity, and replication configuration information.
Note that if you are managing several grids, or a large number of repositories on one grid, you can use the thumbnail images at the bottom of the panel to gain a clearer view. Select a grid by clicking the grid thumbnail image, and a larger view is displayed.
Now that your replication grid and repositories are configured for your VTL systems, you are ready to replicate. To replicate, you need to create policies. Replication policies help you manage when replication occurs, the order in which it occurs, and which information is replicated. You could, for example, ensure that critical data is replicated before nonessential data.
To create a replication policy, click Systems Management to return to the ProtecTIER Manager interface, and select the Replication Policy button. The Create policy wizard opens. Enter a meaningful name for the policy and select whether to enable the policy after it is created. In the next panel, select the destination repository.
Select whether to enable visibility. Visibility controls the virtual location of the replicated cartridges at the destination repository. If you would like the replicated cartridges to be visible to the backup application, enable replication, and select the destination repository and library. If you would like the replicated cartridges to be invisible to the backup application, do not enable visibility, and a shelf will be the repository destination for all cartridges.
Next, enter the barcode ranges for the cartridges that you are replicating, and click Add. You can enter more than one range. Note that you cannot include a cartridge barcode in more than one policy.
Finally, review your selections to complete the wizard.
Your policy is created and displayed in the Replication Policies view. In the right pane, you can view details about a selected policy.
To modify replication policies you have created, right-click the policy, and select Modify replication policy. Use the Modify replication policy wizard to edit barcode ranges or make other changes.
To manually run the policy, right-click the policy, and select Run replication policy. The policy runs automatically when backup is running, depending on the priority level that was set.
You can control when replication occurs by selecting Set replication timeframe. The Prioritized backup option is selected by default. With this option, no specific time frame is set, so replication occurs according to the priority level, and begins automatically when backup activities begin. Select the No backup priority option to set a specific replication time frame. A schedule is displayed, which shows the days of the week and hours of the day. Select the time frame direction and then click areas on the schedule to define half-hour intervals for replication activities. In this example, replication will occur at 2:30 a.m. and 5:30 p.m. on every day except Saturday.
When a replication policy is running, you can monitor progress by selecting Activities from the left navigation pane. Here, you will see the cartridge barcode, the associated policy, the nominal throughput rate, the deduplication factoring ratio, a progress bar, and the time remaining for the replication to complete. In the Overall statistics pane, you can monitor incoming and outgoing replication activity. This pane reflects a many-to-one replication configuration. Both physical and nominal data are shown. For example, if you are replicating 100 gigabytes of data that has been deduplicated to become 10 gigabytes of data, then the nominal amount would be 100 gigabytes and the physical amount would be 10 gigabytes. You also see the backlog, which represents the replication backlog for all cartridges on the repository, not just the cartridges identified in the policy that is running.
If the local or remote site contains a cluster of two ProtecTIER nodes, and one of those nodes fails, then the other node continues with replication activity. This failover is nondisruptive to replication activity. However, if there is only one node at a local or remote site, and it fails, then replication activity does not occur.
If you encounter a disaster or situation where a remote VTL repository goes offline, you can perform disaster recovery to restore data continuity when it comes back online. There are several disaster-recovery stages.
First, select Replication > Replication Disaster Recovery > Enter DR mode and select the offline repository from the list. This ensures that all incoming replication activities from the offline repository are blocked.
After the failed repository is rebuilt or replaced, you can return the data to the site through the failback process.
To launch the Failback policy wizard, select Replication > Replication Disaster Recovery > Fail back. Enter a policy name, select the repository that failed, select whether to enable visibility, specify the barcode ranges, view the validation information, and click Finish.
The failback policy runs. You can view progress in the Grid panel, which indicates the direction of data flow.
To help with restoring data, you can view the replica properties for the cartridges to see when a specific cartridge was last in sync with the original cartridge. This information can help you determine the cartridges that are valid for restoration. Another way to make this determination is to select Show backlog by source and identify the replication group member with the most recent backlog. You can also use the Cartridge Query Tool to generate a statistics report, which lists the cartridges that were replicated.
Next, keep in mind that replicated cartridges at the remote repository are in read-only mode. If you need to run local backups at the target site, you must use new, locally created cartridges.
Finally, you can leave disaster-recovery mode to resume normal backup and replication operations. Select Replication > Replication Disaster Recovery > Leave DR mode, select the repository, and click Ok. Normal replication activity automatically resumes.
In this e-Learning module, you learned about replication for VTL repositories, how to perform several important replication tasks, and the steps for disaster recovery operations.
To learn more about replication and disaster recovery, see the IBM ProtecTIER documentation and other topics in this information center.