6.9.2 File Systems
February 16th, 2026
RBAC required: bear: Physical storage configuration
BEAR provides tools for easily creating and managing file systems on devices to cover the following scenarios:
- Creating a new file system on a disk or RAID device
- Replacing a filesystem or directory on a file system with a new disk or RAID device
- Moving a filesystem on a disk or RAID device back to the main system storage area
- Mounting an filesystem from a an attached disk
BEAR exclusively uses the XFS file system. Any data on partitions created on other file systems will need to be managed directly at the ACHE level by the system administrator. Non XFS file systems will appear in BEAR but you will not be able to manage them.
This article covers the following:
6.9.2.1 About Block Sizes
In general, the block size determines the amount of IO and storage efficiency. Small block sizes incur more IO but are able to more efficiently pack file data. Large block sizes reduce the number of IO ops but may result in wasted space when storing large numbers of small files.
| Use Case |
Recommended Block Size |
Compatible Sector size |
Notes |
| Multimedia files |
4 KiB |
512 b, 1 KiB, 2 KiB, or 4 KiB |
Extremely large files with high throughput requirements will benefit from reading/writing larger blocks at a time. |
| Database datafiles |
512 b, 1 KiB, 2 KiB, or 4 KiB |
4 KiB |
Extremely large files with high IO requirements across the datafiles will benefit from reading/writing larger blocks at a time. |
| Small files |
1 KiB |
512 b, or 1 KiB |
Large numbers of very small files will both store more efficienty with the smaller 1 KiB file boundary resulting in less wasted space and may read faster if they are under 4 KiB in size on average. |
| Mixed use |
2 KiB |
512 b, 1 KiB, or 2 KiB |
A non dedicated use case could have a mixture of many tiny files and large datafiles. 2 KiB provides a reasonable compromise between the 1 Kib and 4 KiB options. |
Regardless of the recommendations above, the specifications of the disk may effectively make the decision for you.
6.9.2.2 Creating a New File System
Before You Begin
- Ensure that the physical storage devices for the file system are attached and, if intended for use on RAID, that the RAID device has also been created. You can only create a new file system if there is at lease one unused physical disk or RAID device.
- Decide on an appropriate block size for the file system. Block devices should be multiples of the disk's physical sector size or, if not possible, at least multiples of the logical sector size. Both of these should be printed somewhere on the physical disk. In addition, the block size should reflect the intended usage. Your application documentation may provide specific guidelines.
File System Creation Steps
- From the Admin Tools menu, select Physical Storage and activate the Create File System tab.

- From the Device dropdown, select a single unused storage device. The fils system will be created to the same size as the selected device.
- Provide an absolute mount path for the file system. This can be an existing directory path or a new directory path.
- Set If mount path contains data to a suitable option. This effectively allows you to migrate data from a file system to a new device that will be mounted at the same location. That operation may incur a service outage for the end user while the data migration is ongoing. You should choose "Move it to the mounted device" if:
- You are using this feature to move application data to a bigger file system.
- Downsizing an oversized file system.
- Replacing the file system with one that is more suitably configured in terms of block size and/or XFS log device.
Creating a file system over an existing directory and moving the data is functionally identical to using the Migrate feature.
- Select the Block size.
- If desired, select an unused disk as a XFS log device.
Log devices should be on disks that perform as good as or better than the main disks used for the file system. Otherwise performance may suffer.

- Click Create File System to start the file system creation job.
XFS Log devices and block sizes cannot be changed after a file system is created.
6.9.2.3 Migrating an Existing File System
This can be done towards a new unused storage device by following the Creating a New File System procedure described above or using the Migrate feature. For unused storage, these are identical. However, the Mirgate feature also also reverse migration of data on a file system with dedicated storage back onto the main system storage pool.
Before You Begin
- If migrating data back to the main storage pool, ensure that the pool (file system /) has enough free space.
- Ensure that the physical storage devices for the file system are attached and, if intended for use on RAID, that the RAID device has also been created. You can only create a new file system if there is at lease one unused physical disk or RAID device.
- Decide on an appropriate block size for the file system. Block devices should be multiples of the disk's physical sector size or, if not possible, at least multiples of the logical sector size. Both of these should be printed somewhere on the physical disk. In addition, the block size should reflect the intended usage. Your application documentation may provide specific guidelines.
File System Migration Steps
- From the Admin Tools menu, select Physical Storage and activate the Migrate tab.

- Select the Source file system from the dropdown list and set the first Target device. There are two options:
- Unused Device: Select this to move the file system to a new unused disk or RAID device.
- Main system disk: Select this to move data off a physical disk or RAID device and onto the main system storage area.

- [Unused device only] From the second Target device dropdown, select the actual unused disk or RAID array to move the filesystem to, set a Block size and, if desired, set an XFS log device.

- Click Migrate to start the file system migration job.
6.9.2.4 Mounting an Existing File System Partition
In some situations, it may be desirable to mount an existing XFS partition from a disk or RAID device. This is useful for the purpose of moving disks between hosts and may be required as part of documented migration steps for an application.
BEAR will automatically detect list all available but unmounted partitions for disks attached to the host. Mounting a partition is performed as follows:
- From the Admin Tools menu, select Physical Storage and activate the Mount tab.

- From the Partition dropdown list, select the partition you want to mount.
- Provide a mount path for the partition. This should not exist on the host.
- Select the mount options. Two are available:
- Persistent (default): Causes the partition to stay mounted. If this is not selected then the partition will be unmounted at the next system reboot.
- Read only: Whether the partition should be mounted in read-only mode to prevent changes being made to the data.

- Click Mount file system to mount the file system and make the data on the partition accessible.
6.9.2.5 Defragmenting a File System
File systems which are read heavy but infrequently written to will tend to have low fragmentation. Whereas file systems that undergo constant change will have a fragmentation ratio that increases relatively quickly.
However, the need to actually defragment a file system may depend on several considerations:
- Has performance noticeably degraded and are high IO service times to blame?
- Is the disk a physical hard disk or a solid state device of some type? SSD drives will generally not be impacted much, if at all, by high fragmentation.
- While high fragmentation on spinning disks (HDD) will affect file system service times, running a defrag job will further impact performance until the job completes. On large file systems, a defrag job could take days or even weeks to complete.
Defragmentation Steps
- From the Admin Tools menu, select Physical Storage and then activate the Defragment & Repair tab.

- Click Analyse File Systems to generate the fragmentation report. This should only take a few minutes.

- Select all the file systems that you wish to defragment and click Run Defrag. The background defrag job will start.
