When you configure the filer volume containing your Exchange data as a SnapMirror source volume, every time a SnapManager backup is taken, the SnapMirror destination volume is optionally updated.
Taken alone, this schedule might not provide an optimal SnapMirror replication schedule. This is because any changes that occur between replications are not replicated to the destination volume until the next replication update. Any restore from the destination volume would restore the Exchange data to its state at the last backup time; subsequent changes to the Exchange data store would not be restored.
Rather than increasing the frequency of SnapManager backups, you can use the rolling snapshot capability of SnapDrive to provide frequent replication of the LUNs that contain the Exchange transaction logs. With more current transaction logs, a restore from the destination volume replays the transaction logs, resulting in an up-to-the-minute restore for the time of the last transaction log update. This method enables you to keep your destination volume more current without running the risk of overlapping your SnapManager operations.
When you use the SnapDrive update mirror, special snapshots called rolling snapshots are used. These snapshots are created only for SnapMirror replication. For more information about rolling snapshots, see your SnapDrive documentation.
There are three major advantages to using SnapDrive rolling snapshots to augment your SnapMirror replication schedule for SnapManager:
When you use SnapDrive to initiate frequent updates to the transaction logs, a restore from the destination volume replays the transaction logs, resulting in an up-to-the-minute restore for the time of the last transaction log update.
It is not recommended to take SnapManager backups every few minutes. This can result in overlapping SnapManager operations and more difficulty managing the large number of resulting backups.
A maximum of two rolling snapshots are retained at any time.
To create a SnapMirror replication schedule for your SnapManager backups using both SnapManager and the SnapDrive command-line interface tool (sdcli), complete the following steps.
| 1 |
Configure a mirror on the filer for the volume occupied by the LUNs backed up by SnapManager. Create an empty replication schedule ("- - - -") where you would normally enter the schedule parameters. For more information about setting a SnapMirror replication schedule, see your SnapDrive documentation and the Data ONTAP Data Protection: Online Backup and Recovery Guide. |
| 2 |
Schedule your SnapManager backups according to your local requirements using either the SnapManager Backup Wizard or the Backup tab. For more information about SnapManager backup schedules, see Recommendations for scheduling backups and verifications. For more information about setting up scheduled SnapManager backups, see Performing a backup. |
| 3 |
Add the sdcli command to the Windows Task Scheduler using the appropriate command-line switches. Take the following requirements into account:
ExampleThe following command can be scheduled successfully on the local host, assuming that Z: is the drive where the transaction logs reside: sdcli snap update_mirror -d Z For more information about sdcli, see your SnapDrive documentation. |
The following table shows an example of a schedule documenting the additional intervals when the volume containing the transaction logs is replicated. The schedule shown in this table allows for a thirty-minute elapse time between the SnapManager backup and the transaction logs drive replication.
You might need to adjust the timings in this example for your environment to account for these requirements:
Copyright © 1994-2004, Network Appliance, Inc. All rights
reserved.
Part No. 215-00858
File: mirror4.htm
Last updated: Tue Jun 15 15:34:13 2004