Disks can be Hard Disk Drives(HDDs), Solid State Drives(SSDs), USB drives, virtual disks from the hypervisor etc.. depending on your setup. Rockstor supports whole disks, but not partitions.
Disk related operations can be managed from the Disks screen listed under the Storage tab of the Web-UI. At a first glance, an overview of disks currently or previously known to the system are displayed.
Table links from left to right:-
- Drive Name - see that drives SMART data / status
- Bulb Icon - to flash the drive’s activity light ie to identify its location
- Pool Name - that pools specific details page
- S.M.A.R.T - enable or disable for each device
- Rescan - the hardware for any supported drives, see Scan for new disks below
- S.M.A.R.T - system wide custom configuration (advanced)
The disks table can be sorted by individual columns by clicking the small up/down arrows displayed in each column header.
With a large numbers of disks the table will be paginated and the current page number will be displayed below the table along with Prev and Next buttons.
Scan for new disks¶
In the disksoverview (above) clicking the Rescan button starts a scan for any new disks added to the system since the last scan. This is particularly useful if a drive has been added to the system since Rockstor was power-on ie hot plugged.
Existing whole disk BTRFS¶
If after scanning or after a Rockstor reinstall the system finds an existing whole disk BTRFS filesystems a small down arrow icon next to the drive name will indicate this. This down arrow can be used to import the discovered btrfs filesystem. In this situation there are two options:
- wipe the disk and re-use as if new - see Wiping a Partitioned Disk below
Rockstor works only with whole disk drives that do not contain a partition table. If a disk has a partition table, it is suspected to have data and Rockstor doesn’t allow it’s usage until the partition table is explicitly wiped. Such disks are displayed with a little eraser icon next to their name. Relevant help text is displayed upon mousing over this icon indicating the above restriction. See the next section Wiping a Partitioned Disk for the procedure.
Wiping a Partitioned Disk¶
Before a partitioned or previously used disk can be deployed it’s partition table or existing whole disk filesystem needs to be wiped, as indicated by the help text above. Click on the eraser icon and a popup confirmation dialog will be displayed. Upon confirmation, the entire disk will be wiped and ready for use as shown below.
Please note that whole disk btrfs filesystems can also be imported, see: Existing whole disk BTRFS
Broken or removed disks¶
Rockstor detects when a disk drive goes offline (damaged or removed from the system) and marks it as such. This is indicated by a little trash icon next to the disk name and relevant help text is displayed upon mousing over this icon.
In order to remove the disk from Rockstor click on the trash icon and a popup confirmation dialog is displayed. Upon confirmation, the disk will be removed as shown below.
A Pool in Rockstor is a set of disk drives combined and represented as a single volume. Pools have attributes such as redundancy profile and compression to safeguard and store data efficiently. Pools can be expanded or shrunk by adding or removing disk drives. In other words, a Pool is a single or multi device BTRFS filesystem.
Pool related operations can be managed from the Pools screen listed under the Storage tab of the Web-UI.
Creating a Pool¶
Only whole disk drives can be used to create Pools. But they don’t have to be of the same size, which is a great feature of BTRFS. Disks that are partitioned with other filesystems including BTRFS won’t be touched. Whole disks with BTRFS created outside of Rockstor or from a previous install, can however be imported. For more information, see Data Import.
Click on Create Pool button and submit the create pool form to create a pool. There is a tooltip for each input field to help you choose appropriate parameters. Here’s a video showing this operation.
All standard BTRFS redundancy profiles are available when creating a pool.
- Single: This profile offers no redundancy, and is the only valid option for creating a Pool with a single disk drive. It is also recommened if you have multiple disks of different sizes, yielding higher total capacity compared to Raid0. Data is neither mirrored nor striped, so if a disk fails, the entire data of the Pool will be lost.
- Raid0: Two or more disks can be used with this profile when there is no need for redundancy. Both metadata and data are striped across the disks. It is recommended for same size disks. If you have different size disks and no need for redundancy, Single profile provides higher capacity. If a disk fails, the entire data of the Pool will be lost.
- Raid1: Two or more disks can be used with this profile. And both metadata and data are replicated on 2 devices. So a Pool with this profile can sustain a single disk failure.
- Raid5: Two or more disks can be used with this profile, which supports striping + parity. Like Raid1, this profile can sustain a single disk failure. The BTRFS community consensus is that Raid5 support is not yet fully stable.
- Raid6: Three or more disks can be used with this profile, which supports striping + dual-parity. Because of dual-parity, a Raid6 Pool can sustain upto two disk failures at the same time. The BTRFS community consensus is that Raid6 support is not yet fully stable.
- Raid10: This is a Raid0 of Raid1 mirrors, with a minimum requirement of four disk drives. It offers most redundancy at the cost of capacity where a Pool can sustain multiple disk failures at the same time as long as they part of different Raid1 groups.
Compression can optionally be chosen during Pool creation or it can be set on a previously created Pool. In the latter scenario, compression is applied only to data written after it’s set.
Compression can also be set at the Share level. If you don’t want to enable compression for all Shares under a Pool, don’t enable it at the Pool level. Instead, selectively enable it on Shares.
Besides not enabling compression at all, there are two additional choices
These are optional and meant for more advanced users to provide BTRFS specific mount options. Since a Pool is a filesystem, it is mounted with default options which can be altered by providing one or more of the following. You can find out more about each option here.
Resizing a Pool¶
You can resize a Pool for one of the following reasons
- To change it’s redundancy profile. For example, to go from a RAID10 to RAID1.
- To add more disks and increase it’s capacity.
- To remove disks and decrease capacity. Removed disks can be reused for other Pools.
Pool resize is an online operation that does not cause any access disruption. However, depending on size of the Pool, it could take a long time to finish. Resizing is simple and can be done using the Web-UI.
This video shows changing the redundancy profile
This video shows adding disks to a Pool
This video shows removing disks from a Pool
Deleting a Pool¶
A pool can be deleted as long as it is empty, i.e., there are no shares remaining in it.
Go to the Storage tab of the Web-UI and click on Pools in the left sidebar to enter the Pools view. In the displayed table of pools, click on the trash icon corresponding to the pool to delete it as shown below.
A pool can also be deleted by clicking the Delete button inside it’s detail view.
Scrubbing a Pool¶
Over time, a pool could accumulate low level errors relating to redundancy. Scrubbing is a background process that finds and fixes these errors and ensures the long life of a pool.
The pool scrub operation can take a while depending on the size of the pool. To start a scrub, go to the pool’s detail view and click on the Start a new scrub button under the scrub tab. The button will be disabled during the scrub process and enabled again once the scrub finishes.
Balance a pool¶
click on Start a new balance. [NOTE : Suman to complete the documentation for balances]