Storage management

You can create and manage storage pools for a vSnap server. You can also manage the cache and the log files for the server.

Managing disks

The vSnap server creates a storage pool by using the disks that are provisioned to the vSnap server. In the case of virtual deployments, the disks can be RDM or virtual disks provisioned from datastores on any backing storage. In the case of physical deployments, the disks can be local or be attached to the physical server in a storage area network (SAN). The local disks might already have external redundancy enabled via a hardware Redundant Array of Independent Disks (RAID) controller, but if not, the vSnap server can create RAID-based storage pools for internal redundancy.

Attention: Disks that are attached to vSnap servers must be thick provisioned. If disks are thin provisioned, the amount of free space in the storage pool might not be adequately reported. This situation might lead to data corruption if the underlying datastore runs out of space.

After a disk is added to a storage pool, do not remove the disk. Removing the disk will corrupt the storage pool.

If the vSnap server was deployed as part of a virtual appliance, the appliance already contains a 100 GB starter virtual disk. For instructions about managing this disk, see the Blueprints. You can add more disks before or after creating a pool and accordingly use them to create a larger pool or expand an existing pool. If job logs report that a vSnap server is reaching its storage capacity, additional disks can be added to the vSnap pool. Or you can create an SLA policy and specify that backup operations use an alternative vSnap server as the target.

You can prevent data corruption, which can occur when a VMware datastore on a vSnap server reaches its capacity. Create a stable environment for virtual vSnap servers that use RAID configurations and utilize thick provisioned VMDKs. By replicating data to external vSnap servers, you can provide additional protection.

A vSnap server will become invalidated if the vSnap pool is deleted or if a vSnap disk is deleted. All data on the vSnap server will be lost. If your vSnap server becomes invalidated, you must unregister the vSnap server by using the IBM Storage® Protect Plus interface, and then run the maintenance job. When the maintenance job is complete, register the vSnap server again.

Enabling encryption

To enable encryption of backup data on a vSnap server, select Initialize with encryption enabled when you initialize the server. Encryption settings cannot be changed after the server is initialized and a pool is created. All disks of a vSnap pool use the same encryption key file, which is generated upon pool creation. Data is encrypted when at rest on the vSnap server.

vSnap encryption utilizes the following algorithm:
Cipher name
Advanced Encryption Standard (AES)
Cipher mode
xts-plain64
Key
256 bits
Linux Unified Key Setup (LUKS) header hashing
sha256

Managing encryption keys

The disk encryption key files that are generated during pool creation are stored under the directory /etc/vsnap/keys/ on each vSnap server. For disaster recovery purposes, back up the key files manually to another location outside of the vSnap server. After a pool is created, use the following commands as the serveradmin user to copy the keys to a temporary location and then copy them to a secure backup location outside the vSnap host. Complete the following steps:
  1. Create a directory to which the keys will be backed up:
    $ mkdir /tmp/keybackup-$(hostname)
  2. Copy the key files to the temporary location:
    $ sudo cp -r /etc/vsnap/keys /tmp/keybackup-$(hostname)
  3. Copy the keybackup-<hostname> directory to a secure backup location outside of the vSnap host.

Detecting disks

If you add disks to a vSnap server, use the command line or the IBM Storage Protect Plus user interface to detect the newly attached disks.

Command line: Run the $ vsnap disk rescan command.

User interface: In the navigation panel, click System Configuration > Storage > vSnap servers. Then, click Manage. Open the Storage disk tab, and then click Rescan.

Showing disks

To view a list of all disks in the vSnap system, run the $ vsnap disk show command.

The USED AS column in the output shows whether each disk is in use. Any disk that is unformatted and unpartitioned is marked as unused. All other disks are marked as used.

Only disks that are marked as unused can be used to create a storage pool or be added to a storage pool. If a disk that you plan to add to a storage pool is not marked as unused, it might be because the disk was previously in use and thus contains remnants of an older partition table or file system. You can correct this issue by using system commands like parted or dd to wipe the disk partition table.

Showing storage pool information

To view information about each storage pool, run the $ vsnap pool show command.

Creating a storage pool

If you completed the simple initialization procedure that is described in Completing a simple initialization, a storage pool was created automatically and the information in this section is not applicable.

To complete an advanced initialization, use the vsnap pool create command to create a storage pool manually. Before you run the command, ensure that one or more unused disks are available as described in Showing disks. For information about available options, use the --help option for any command or subcommand.

Specify a display name for the pool and a list of one or more disks. If no disks are specified, all available unused disks are used. You can enable compression and deduplication for the pool during creation. You can also update the compression and deduplication settings later by using the vsnap pool update command.

The pool type that you specify during the creation of the storage pool specifies the redundancy of the pool:
raid0
This is the default option when no pool type is specified. If this option is used, vSnap assumes that your disks have external redundancy. This setting is appropriate, for example, if you use virtual disks on a datastore that is backed by redundant storage. In this case, the storage pool has no internal redundancy.
After a disk is added to a raid0 pool, the disk cannot be removed. If you remove the disk, the pool becomes unavailable. This issue can be resolved only by destroying and re-creating the pool.
raid5
When you select this option, the pool is comprised of one or more RAID5 group, each consisting of three or more disks. The number of RAID5 groups and the number of disks in each group depend on the total number of disks that you specify during pool creation. Based on the number of available disks, vSnap uses values that maximize total capacity while also helping to optimize redundancy of metadata.
raid6
When you select this option, the pool is comprised of one or more RAID6 group, each consisting of four or more disks. The number of RAID6 groups and the number of disks in each group depend on the total number of disks that you specify during pool creation. Based on the number of available disks, vSnap uses values that maximize total capacity while also helping to optimize redundancy of metadata.

Expanding a storage pool

Before you expand a pool, ensure that one or more unused disks are available as described in Showing disks.

Use the command line or the IBM Storage Protect Plus user interface to expand a storage pool.

Command line: Run the $ vsnap pool expand command. For information about available options, use the --help option for any command or subcommand.

User interface: In the navigation panel, click System Configuration > Storage > vSnap servers. Then, click Manage. Open the Storage disk tab. The tab displays all unused disks that are detected on the system. Select one or more disks and click Save to add them to the storage pool.

Managing the cache and log for storage pools

To store cache and log data for vSnap storage, use solid-state drive (SSD) flash or non-volatile memory express (NVMe) disks. By adding cache and log space to storage pools, you can help to optimize the performance of the vSnap server by decreasing redundant input and output (I/O) to the server. For more information about configuring cache and log space for storage pools, see the IBM Storage Protect Plus Blueprints.

You must use the command line to add or remove the cache and log. Because the cache and log do not store data permanently, you can remove them when the pool is online. However, ensure that no backup, restore, or replication operations are occurring before you issue the remove command.

Use the following commands to add and remove the cache or log. For information about the available options for a command, use the --help option. For examples of these commands as used in vSnap installation and configuration steps, see the IBM Storage Protect Plus Blueprints.
Attention: Do not remove the devices that are providing space for the log and cache from the vSnap system without first removing the log and cache from the storage pool by using the appropriate remove command.
  • vsnap pool addcache
  • vsnap pool addlog
  • vsnap pool removecache
  • vsnap pool removelog