Scratch Volume Pools

Topics | How To | Related Topics |


  This feature/product/platform is on Extended Support in this release. See Deprecated Features, Products, and Platforms for more information.

Overview

QR Volumes and Scratch Volume Pools

Scratch Volume Pool Considerations

Recovery and Scratch Volume Pools


Overview

A Scratch Volume Pool is a repository of destination volumes that are available for use by a QR Volume Creation operation. The system does not create Scratch Volume Pools by default. The Pools must be manually created and assigned to a QR Policy.  Scratch Volume Pools are created in the Storage Resources section of the CommCell Browser. 


QR Volumes and Scratch Volume Pools

When creating a QR Volume, you may select a specific destination volume or allow the Quick Recovery Agent to select one automatically. Destination volumes must be selected from the Scratch Volume Pool associated with the QR Policy. The Quick Recovery Agent will select an un-allocated volume from the pool unless a specific volume is chosen by the user. The agent will attempt to allocate the smallest available volume that is of a sufficient size to hold the primary volume's contents. However, it is recommended that you create different Scratch Volume Pools containing volumes of consistent size, and assign the pool of appropriate size to each QR Policy and/or subclient as needed.

A disk volume must be detected and configured by Volume Explorer before it can be assigned to a Scratch Volume Pool. A single volume may only belong to one Scratch Volume Pool at a time. The Quick Recovery Agent keeps track of scratch volume allocations and will prevent multiple QR Volume operations from attempting to use the same scratch volume. If a QR Volume Creation/Update operation fails unexpectedly, it may leave the destination volume in a reserved state. The CommCell Archive Pruning operation will detect and clean up any such volumes the next time it is run.

Only QSnap (CXBF) devices can be used as destination volumes when using QSnap and the Quick Recovery Agent. Otherwise, the destination volume could become inconsistent with the source. If a QSnap (CXBF) device becomes de-configured, it may be chosen from the Scratch Volume Pool as a destination volume. To avoid this, manually choose volumes when creating QR destination Volumes; look for "cxbf" in the Device Name.

Scratch Volume Pool Considerations

When creating a Scratch Volume Pool, you must ensure that:

The user can manually add and delete volumes to and from an existing pool.  Once a copy has been created and data has been copied to a destination volume in your associated scratch volume pool, the volume will be marked locked, which can be seen from the Volume Locked check box in the right panel of the scratch volume pool.  Once this volume is locked, any subsequent QR Volume creation jobs will not be allowed to use it as a destination volume. This is to protect the volume from being overwritten. Any destination volumes which are locked but no longer in use by any QR Volumes will be released during the next data aging job on the CommServe.

You can create any number of additional scratch volume pools and assign them to any QR policies that you wish.  The user can also add volumes from different clients into the same pool, which allows for more flexibility in QR volume creation.  The user cannot, however, overlap volumes. Scratch volume pools are mutually exclusive.  For instance, the F: drive on client1 cannot be used in more than one scratch volume pool.  To prevent an overlap from occurring, available volumes to be added to the scratch volume pool are filtered so that assigned volumes will not be viewable from the list.


Recovery and Scratch Volume Pools

When a recover is performed, the mount point or drive letter of the destination volume is switched with that of the source volume.  In other words,  the scratch volume pool will show the source volume's drive letter in place of the destination volume. This concept can be confusing, so an example is provided below:  

If the user performs a QR Volume creation with F: drive (source) to the Z: drive (destination), then a QR recover will unmount the source (F:) and destination volume (Z:) and mount the QR volume as F:.

This is an automatic feature of Recovery and requires no user intervention.  From this point the user has some options.  The most common choice would be to delete the QR Volume and remove it from the scratch volume pool.   The other option is that the user could choose to leave the recovered volume in a locked status.

Back to Top