Red Hat Training

A Red Hat training course is available for RHEL 8

Chapter 43. Deduplicating and compressing storage

43.1. Deploying VDO

As a system administrator, you can use VDO to create deduplicated and compressed storage pools.

43.1.1. Introduction to VDO

Virtual Data Optimizer (VDO) provides inline data reduction for Linux in the form of deduplication, compression, and thin provisioning. When you set up a VDO volume, you specify a block device on which to construct your VDO volume and the amount of logical storage you plan to present.

  • When hosting active VMs or containers, Red Hat recommends provisioning storage at a 10:1 logical to physical ratio: that is, if you are utilizing 1 TB of physical storage, you would present it as 10 TB of logical storage.
  • For object storage, such as the type provided by Ceph, Red Hat recommends using a 3:1 logical to physical ratio: that is, 1 TB of physical storage would present as 3 TB logical storage.

In either case, you can simply put a file system on top of the logical device presented by VDO and then use it directly or as part of a distributed cloud storage architecture.

Because VDO is thinly provisioned, the file system and applications only see the logical space in use and are not aware of the actual physical space available. Use scripting to monitor the actual available space and generate an alert if use exceeds a threshold: for example, when the VDO volume is 80% full.

43.1.2. VDO deployment scenarios

You can deploy VDO in a variety of ways to provide deduplicated storage for:

  • both block and file access
  • both local and remote storage

Because VDO exposes its deduplicated storage as a standard Linux block device, you can use it with standard file systems, iSCSI and FC target drivers, or as unified storage.

Note

VDO deployment with Ceph Storage is currently not supported.

KVM

You can deploy VDO on a KVM server configured with Direct Attached Storage.

VDO Deployment with KVM
File systems

You can create file systems on top of VDO and expose them to NFS or CIFS users with the NFS server or Samba.

Deduplicated NAS
Placement of VDO on iSCSI

You can export the entirety of the VDO storage target as an iSCSI target to remote iSCSI initiators.

Deduplicated block storage target

When creating a VDO volume on iSCSI, you can place the VDO volume above or below the iSCI layer. Although there are many considerations to be made, some guidelines are provided here to help you select the method that best suits your environment.

When placing the VDO volume on the iSCSI server (target) below the iSCSI layer:

  • The VDO volume is transparent to the initiator, similar to other iSCSI LUNs. Hiding the thin provisioning and space savings from the client makes the appearance of the LUN easier to monitor and maintain.
  • There is decreased network traffic because there are no VDO metadata reads or writes, and read verification for the dedupe advice does not occur across the network.
  • The memory and CPU resources being used on the iSCSI target can result in better performance. For example, the ability to host an increased number of hypervisors because the volume reduction is happening on the iSCSI target.
  • If the client implements encryption on the initiator and there is a VDO volume below the target, you will not realize any space savings.

When placing the VDO volume on the iSCSI client (initiator) above the iSCSI layer:

  • There is a potential for lower network traffic across the network in ASYNC mode if achieving high rates of space savings.
  • You can directly view and control the space savings and monitor usage.
  • If you want to encrypt the data, for example, using dm-crypt, you can implement VDO on top of the crypt and take advantage of space efficiency.
LVM

On more feature-rich systems, you can use LVM to provide multiple logical unit numbers (LUNs) that are all backed by the same deduplicated storage pool.

In the following diagram, the VDO target is registered as a physical volume so that it can be managed by LVM. Multiple logical volumes (LV1 to LV4) are created out of the deduplicated storage pool. In this way, VDO can support multiprotocol unified block or file access to the underlying deduplicated storage pool.

Deduplicated unified storage

Deduplicated unified storage design enables for multiple file systems to collectively use the same deduplication domain through the LVM tools. Also, file systems can take advantage of LVM snapshot, copy-on-write, and shrink or grow features, all on top of VDO.

Encryption

Device Mapper (DM) mechanisms such as DM Crypt are compatible with VDO. Encrypting VDO volumes helps ensure data security, and any file systems above VDO are still deduplicated.

Using VDO with encryption
Important

Applying the encryption layer above VDO results in little if any data deduplication. Encryption makes duplicate blocks different before VDO can deduplicate them.

Always place the encryption layer below VDO.

When creating a VDO volume on iSCSI, you can place the VDO volume above or below the iSCI layer. Although there are many considerations to be made, some guidelines are provided here to help you select the method that best suits your environment.

43.1.3. Components of a VDO volume

VDO uses a block device as a backing store, which can include an aggregation of physical storage consisting of one or more disks, partitions, or even flat files. When a storage management tool creates a VDO volume, VDO reserves volume space for the UDS index and VDO volume. The UDS index and the VDO volume interact together to provide deduplicated block storage.

Figure 43.1. VDO disk organization

VDO disk organization

The VDO solution consists of the following components:

kvdo

A kernel module that loads into the Linux Device Mapper layer provides a deduplicated, compressed, and thinly provisioned block storage volume.

The kvdo module exposes a block device. You can access this block device directly for block storage or present it through a Linux file system, such as XFS or ext4.

When kvdo receives a request to read a logical block of data from a VDO volume, it maps the requested logical block to the underlying physical block and then reads and returns the requested data.

When kvdo receives a request to write a block of data to a VDO volume, it first checks whether the request is a DISCARD or TRIM request or whether the data is uniformly zero. If either of these conditions is true, kvdo updates its block map and acknowledges the request. Otherwise, VDO processes and optimizes the data.

uds

A kernel module that communicates with the Universal Deduplication Service (UDS) index on the volume and analyzes data for duplicates. For each new piece of data, UDS quickly determines if that piece is identical to any previously stored piece of data. If the index finds a match, the storage system can then internally reference the existing item to avoid storing the same information more than once.

The UDS index runs inside the kernel as the uds kernel module.

Command line tools
For configuring and managing optimized storage.

43.1.4. The physical and logical size of a VDO volume

This section describes the physical size, available physical size, and logical size that VDO can utilize:

Physical size

This is the same size as the underlying block device. VDO uses this storage for:

  • User data, which might be deduplicated and compressed
  • VDO metadata, such as the UDS index
Available physical size

This is the portion of the physical size that VDO is able to use for user data

It is equivalent to the physical size minus the size of the metadata, minus the remainder after dividing the volume into slabs by the given slab size.

Logical Size

This is the provisioned size that the VDO volume presents to applications. It is usually larger than the available physical size. If the --vdoLogicalSize option is not specified, then the provisioning of the logical volume is now provisioned to a 1:1 ratio. For example, if a VDO volume is put on top of a 20 GB block device, then 2.5 GB is reserved for the UDS index (if the default index size is used). The remaining 17.5 GB is provided for the VDO metadata and user data. As a result, the available storage to consume is not more than 17.5 GB, and can be less due to metadata that makes up the actual VDO volume.

VDO currently supports any logical size up to 254 times the size of the physical volume with an absolute maximum logical size of 4PB.

Figure 43.2. VDO disk organization

VDO disk organization

In this figure, the VDO deduplicated storage target sits completely on top of the block device, meaning the physical size of the VDO volume is the same size as the underlying block device.

Additional resources

43.1.5. Slab size in VDO

The physical storage of the VDO volume is divided into a number of slabs. Each slab is a contiguous region of the physical space. All of the slabs for a given volume have the same size, which can be any power of 2 multiple of 128 MB up to 32 GB.

The default slab size is 2 GB in order to facilitate evaluating VDO on smaller test systems. A single VDO volume can have up to 8192 slabs. Therefore, in the default configuration with 2 GB slabs, the maximum allowed physical storage is 16 TB. When using 32 GB slabs, the maximum allowed physical storage is 256 TB. VDO always reserves at least one entire slab for metadata, and therefore, the reserved slab cannot be used for storing user data.

Slab size has no effect on the performance of the VDO volume.

Table 43.1. Recommended VDO slab sizes by physical volume size

Physical volume sizeRecommended slab size

10–99 GB

1 GB

100 GB – 1 TB

2 GB

2–256 TB

32 GB

You can control the slab size by providing the --vdoSlabSize=megabytes option to the vdo create command.

43.1.6. VDO requirements

VDO has certain requirements on its placement and your system resources.

43.1.6.1. VDO memory requirements

Each VDO volume has two distinct memory requirements:

The VDO module

VDO requires a fixed 38 MB of RAM and several variable amounts:

  • 1.15 MB of RAM for each 1 MB of configured block map cache size. The block map cache requires a minimum of 150MB RAM.
  • 1.6 MB of RAM for each 1 TB of logical space.
  • 268 MB of RAM for each 1 TB of physical storage managed by the volume.

The UDS index

The Universal Deduplication Service (UDS) requires a minimum of 250 MB of RAM, which is also the default amount that deduplication uses. You can configure the value when formatting a VDO volume, because the value also affects the amount of storage that the index needs.

The memory required for the UDS index is determined by the index type and the required size of the deduplication window:

Index typeDeduplication windowNote

Dense

1 TB per 1 GB of RAM

A 1 GB dense index is generally sufficient for up to 4 TB of physical storage.

Sparse

10 TB per 1 GB of RAM

A 1 GB sparse index is generally sufficient for up to 40 TB of physical storage.

The UDS Sparse Indexing feature is the recommended mode for VDO. It relies on the temporal locality of data and attempts to retain only the most relevant index entries in memory. With the sparse index, UDS can maintain a deduplication window that is ten times larger than with dense, while using the same amount of memory.

Although the sparse index provides the greatest coverage, the dense index provides more deduplication advice. For most workloads, given the same amount of memory, the difference in deduplication rates between dense and sparse indexes is negligible.

Additional resources

43.1.6.2. VDO storage space requirements

You can configure a VDO volume to use up to 256 TB of physical storage. Only a certain part of the physical storage is usable to store data. This section provides the calculations to determine the usable size of a VDO-managed volume.

VDO requires storage for two types of VDO metadata and for the UDS index:

  • The first type of VDO metadata uses approximately 1 MB for each 4 GB of physical storage plus an additional 1 MB per slab.
  • The second type of VDO metadata consumes approximately 1.25 MB for each 1 GB of logical storage, rounded up to the nearest slab.
  • The amount of storage required for the UDS index depends on the type of index and the amount of RAM allocated to the index. For each 1 GB of RAM, a dense UDS index uses 17 GB of storage, and a sparse UDS index will use 170 GB of storage.

Additional resources

43.1.6.3. Placement of VDO in the storage stack

You should place certain storage layers under VDO and others above VDO.

A VDO volume is a thinly provisioned block device. To prevent running out of physical space, place the volume on top of storage that you can expand at a later time. Examples of such expandable storage are LVM volumes or MD RAID arrays.

You can place thick-provisioned layers on top of VDO, but you cannot rely on the guarantees of thick provisioning in that case. Because the VDO layer is thin-provisioned, the effects of thin provisioning apply to all layers above it. If you do not monitor the VDO device, you might run out of physical space on thick-provisioned volumes above VDO.

Supported configurations

  • Layers that you can place only under VDO:

    • DM Multipath
    • DM Crypt
    • Software RAID (LVM or MD RAID)
  • Layers that you can place only above VDO:

    • LVM cache
    • LVM snapshots
    • LVM thin provisioning

Unsupported configurations

  • VDO on top of other VDO volumes
  • VDO on top of LVM snapshots
  • VDO on top of LVM cache
  • VDO on top of a loopback device
  • VDO on top of LVM thin provisioning
  • Encrypted volumes on top of VDO
  • Partitions on a VDO volume
  • RAID, such as LVM RAID, MD RAID, or any other type, on top of a VDO volume

Additional resources

43.1.6.4. Examples of VDO requirements by physical size

The following tables provide approximate system requirements of VDO based on the physical size of the underlying volume. Each table lists requirements appropriate to the intended deployment, such as primary storage or backup storage.

The exact numbers depend on your configuration of the VDO volume.

Primary storage deployment

In the primary storage case, the UDS index is between 0.01% to 25% the size of the physical size.

Table 43.2. Storage and memory requirements for primary storage

Physical sizeRAM usage: UDSRAM usage: VDODisk usageIndex type

10GB–1TB

250MB

472MB

2.5GB

Dense

2–10TB

1GB

3GB

10GB

Dense

250MB

22GB

Sparse

11–50TB

2GB

14GB

170GB

Sparse

51–100TB

3GB

27GB

255GB

Sparse

101–256TB

12GB

69GB

1020GB

Sparse

Backup storage deployment

In the backup storage case, the UDS index covers the size of the backup set but is not bigger than the physical size. If you expect the backup set or the physical size to grow in the future, factor this into the index size.

Table 43.3. Storage and memory requirements for backup storage

Physical sizeRAM usage: UDSRAM usage: VDODisk usageIndex type

10GB–1TB

250MB

472MB

2.5 GB

Dense

2–10TB

2GB

3GB

170GB

Sparse

11–50TB

10GB

14GB

850GB

Sparse

51–100TB

20GB

27GB

1700GB

Sparse

101–256TB

26GB

69GB

3400GB

Sparse

43.1.7. Installing VDO

This procedure installs software necessary to create, mount, and manage VDO volumes.

Procedure

  • Install the vdo and kmod-kvdo packages:

    # yum install vdo kmod-kvdo

43.1.8. Creating a VDO volume

This procedure creates a VDO volume on a block device.

Prerequisites

Procedure

In all the following steps, replace vdo-name with the identifier you want to use for your VDO volume; for example, vdo1. You must use a different name and device for each instance of VDO on the system.

  1. Find a persistent name for the block device where you want to create the VDO volume. For more information on persistent names, see Chapter 32, Overview of persistent naming attributes.

    If you use a non-persistent device name, then VDO might fail to start properly in the future if the device name changes.

  2. Create the VDO volume:

    # vdo create \
          --name=vdo-name \
          --device=block-device \
          --vdoLogicalSize=logical-size
    • Replace block-device with the persistent name of the block device where you want to create the VDO volume. For example, /dev/disk/by-id/scsi-3600508b1001c264ad2af21e903ad031f.
    • Replace logical-size with the amount of logical storage that the VDO volume should present:

      • For active VMs or container storage, use logical size that is ten times the physical size of your block device. For example, if your block device is 1TB in size, use 10T here.
      • For object storage, use logical size that is three times the physical size of your block device. For example, if your block device is 1TB in size, use 3T here.
    • If the physical block device is larger than 16TiB, add the --vdoSlabSize=32G option to increase the slab size on the volume to 32GiB.

      Using the default slab size of 2GiB on block devices larger than 16TiB results in the vdo create command failing with the following error:

      vdo: ERROR - vdoformat: formatVDO failed on '/dev/device': VDO Status: Exceeds maximum number of slabs supported

    Example 43.1. Creating VDO for container storage

    For example, to create a VDO volume for container storage on a 1TB block device, you might use:

    # vdo create \
          --name=vdo1 \
          --device=/dev/disk/by-id/scsi-3600508b1001c264ad2af21e903ad031f \
          --vdoLogicalSize=10T
    Important

    If a failure occurs when creating the VDO volume, remove the volume to clean up. See Removing an unsuccessfully created VDO volume for details.

  3. Create a file system on top of the VDO volume:

    • For the XFS file system:

      # mkfs.xfs -K /dev/mapper/vdo-name
    • For the ext4 file system:

      # mkfs.ext4 -E nodiscard /dev/mapper/vdo-name
  4. Use the following command to wait for the system to register the new device node:

    # udevadm settle

Next steps

  1. Mount the file system. See Section 43.1.9, “Mounting a VDO volume” for details.
  2. Enable the discard feature for the file system on your VDO device. See Section 43.1.10, “Enabling periodic block discard” for details.

Additional resources

  • The vdo(8) man page

43.1.9. Mounting a VDO volume

This procedure mounts a file system on a VDO volume, either manually or persistently.

Prerequisites

Procedure

  • To mount the file system on the VDO volume manually, use:

    # mount /dev/mapper/vdo-name mount-point
  • To configure the file system to mount automatically at boot, add a line to the /etc/fstab file:

    • For the XFS file system:

      /dev/mapper/vdo-name mount-point xfs defaults,x-systemd.device-timeout=0,x-systemd.requires=vdo.service 0 0
    • For the ext4 file system:

      /dev/mapper/vdo-name mount-point ext4 defaults,x-systemd.device-timeout=0,x-systemd.requires=vdo.service 0 0

    If the VDO volume is located on a block device that requires network, such as iSCSI, add the _netdev mount option.

Additional resources

  • The vdo(8) man page.
  • For iSCSI and other block devices requiring network, see the systemd.mount(5) man page for information on the _netdev mount option.

43.1.10. Enabling periodic block discard

This procedure enables a systemd timer that regularly discards unused blocks on all supported file systems.

Procedure

  • Enable and start the systemd timer:

    # systemctl enable --now fstrim.timer

43.1.11. Monitoring VDO

This procedure describes how to obtain usage and efficiency information from a VDO volume.

Prerequisites

Procedure

  • Use the vdostats utility to get information about a VDO volume:

    # vdostats --human-readable
    
    Device                   1K-blocks    Used     Available    Use%    Space saving%
    /dev/mapper/node1osd1    926.5G       21.0G    905.5G       2%      73%
    /dev/mapper/node1osd2    926.5G       28.2G    898.3G       3%      64%

Additional resources

  • The vdostats(8) man page.

43.2. Maintaining VDO

After deploying a VDO volume, you can perform certain tasks to maintain or optimize it. Some of the following tasks are required for the correct functioning of VDO volumes.

Prerequisites

43.2.1. Managing free space on VDO volumes

VDO is a thinly provisioned block storage target. Because of that, you must actively monitor and manage space usage on VDO volumes.

43.2.1.1. The physical and logical size of a VDO volume

This section describes the physical size, available physical size, and logical size that VDO can utilize:

Physical size

This is the same size as the underlying block device. VDO uses this storage for:

  • User data, which might be deduplicated and compressed
  • VDO metadata, such as the UDS index
Available physical size

This is the portion of the physical size that VDO is able to use for user data

It is equivalent to the physical size minus the size of the metadata, minus the remainder after dividing the volume into slabs by the given slab size.

Logical Size

This is the provisioned size that the VDO volume presents to applications. It is usually larger than the available physical size. If the --vdoLogicalSize option is not specified, then the provisioning of the logical volume is now provisioned to a 1:1 ratio. For example, if a VDO volume is put on top of a 20 GB block device, then 2.5 GB is reserved for the UDS index (if the default index size is used). The remaining 17.5 GB is provided for the VDO metadata and user data. As a result, the available storage to consume is not more than 17.5 GB, and can be less due to metadata that makes up the actual VDO volume.

VDO currently supports any logical size up to 254 times the size of the physical volume with an absolute maximum logical size of 4PB.

Figure 43.3. VDO disk organization

VDO disk organization

In this figure, the VDO deduplicated storage target sits completely on top of the block device, meaning the physical size of the VDO volume is the same size as the underlying block device.

Additional resources

43.2.1.2. Thin provisioning in VDO

VDO is a thinly provisioned block storage target. The amount of physical space that a VDO volume uses might differ from the size of the volume that is presented to users of the storage. You can make use of this disparity to save on storage costs.

Out-of-space conditions

Take care to avoid unexpectedly running out of storage space, if the data written does not achieve the expected rate of optimization.

Whenever the number of logical blocks (virtual storage) exceeds the number of physical blocks (actual storage), it becomes possible for file systems and applications to unexpectedly run out of space. For that reason, storage systems using VDO must provide you with a way of monitoring the size of the free pool on the VDO volume.

You can determine the size of this free pool by using the vdostats utility. The default output of this utility lists information for all running VDO volumes in a format similar to the Linux df utility. For example:

Device                1K-blocks   Used        Available   Use%
/dev/mapper/vdo-name  211812352   105906176   105906176   50%

When the physical storage capacity of a VDO volume is almost full, VDO reports a warning in the system log, similar to the following:

Oct  2 17:13:39 system lvm[13863]: Monitoring VDO pool vdo-name.
Oct  2 17:27:39 system lvm[13863]: WARNING: VDO pool vdo-name is now 80.69% full.
Oct  2 17:28:19 system lvm[13863]: WARNING: VDO pool vdo-name is now 85.25% full.
Oct  2 17:29:39 system lvm[13863]: WARNING: VDO pool vdo-name is now 90.64% full.
Oct  2 17:30:29 system lvm[13863]: WARNING: VDO pool vdo-name is now 96.07% full.
Note

These warning messages appear only when the lvm2-monitor service is running. It is enabled by default.

How to prevent out-of-space conditions

If the size of free pool drops below a certain level, you can take action by:

  • Deleting data. This reclaims space whenever the deleted data is not duplicated. Deleting data frees the space only after discards are issued.
  • Adding physical storage
Important

Monitor physical space on your VDO volumes to prevent out-of-space situations. Running out of physical blocks might result in losing recently written, unacknowledged data on the VDO volume.

Thin provisioning and the TRIM and DISCARD commands

To benefit from the storage savings of thin provisioning, the physical storage layer needs to know when data is deleted. File systems that work with thinly provisioned storage send TRIM or DISCARD commands to inform the storage system when a logical block is no longer required.

Several methods of sending the TRIM or DISCARD commands are available:

  • With the discard mount option, the file systems can send these commands whenever a block is deleted.
  • You can send the commands in a controlled manner by using utilities such as fstrim. These utilities tell the file system to detect which logical blocks are unused and send the information to the storage system in the form of a TRIM or DISCARD command.

The need to use TRIM or DISCARD on unused blocks is not unique to VDO. Any thinly provisioned storage system has the same challenge.

43.2.1.3. Monitoring VDO

This procedure describes how to obtain usage and efficiency information from a VDO volume.

Prerequisites

Procedure

  • Use the vdostats utility to get information about a VDO volume:

    # vdostats --human-readable
    
    Device                   1K-blocks    Used     Available    Use%    Space saving%
    /dev/mapper/node1osd1    926.5G       21.0G    905.5G       2%      73%
    /dev/mapper/node1osd2    926.5G       28.2G    898.3G       3%      64%

Additional resources

  • The vdostats(8) man page.

43.2.1.4. Reclaiming space for VDO on file systems

This procedure reclaims storage space on a VDO volume that hosts a file system.

VDO cannot reclaim space unless file systems communicate that blocks are free using the DISCARD, TRIM, or UNMAP commands.

Procedure

  • If the file system on your VDO volume supports discard operations, enable them. See Discarding unused blocks.
  • For file systems that do not use DISCARD, TRIM, or UNMAP, you can manually reclaim free space. Store a file consisting of binary zeros to fill the free space and then delete that file.

43.2.1.5. Reclaiming space for VDO without a file system

This procedure reclaims storage space on a VDO volume that is used as a block storage target without a file system.

Procedure

  • Use the blkdiscard utility.

    For example, a single VDO volume can be carved up into multiple subvolumes by deploying LVM on top of it. Before deprovisioning a logical volume, use the blkdiscard utility to free the space previously used by that logical volume.

    LVM supports the REQ_DISCARD command and forwards the requests to VDO at the appropriate logical block addresses in order to free the space. If you use other volume managers, they also need to support REQ_DISCARD, or equivalently, UNMAP for SCSI devices or TRIM for ATA devices.

Additional resources

  • The blkdiscard(8) man page

43.2.1.6. Reclaiming space for VDO on Fibre Channel or Ethernet network

This procedure reclaims storage space on VDO volumes (or portions of volumes) that are provisioned to hosts on a Fibre Channel storage fabric or an Ethernet network using SCSI target frameworks such as LIO or SCST.

Procedure

  • SCSI initiators can use the UNMAP command to free space on thinly provisioned storage targets, but the SCSI target framework needs to be configured to advertise support for this command. This is typically done by enabling thin provisioning on these volumes.

    Verify support for UNMAP on Linux-based SCSI initiators by running the following command:

    # sg_vpd --page=0xb0 /dev/device

    In the output, verify that the Maximum unmap LBA count value is greater than zero.

43.2.2. Starting or stopping VDO volumes

You can start or stop a given VDO volume, or all VDO volumes, and their associated UDS indexes.

43.2.2.1. Started and activated VDO volumes

During the system boot, the vdo systemd unit automatically starts all VDO devices that are configured as activated.

The vdo systemd unit is installed and enabled by default when the vdo package is installed. This unit automatically runs the vdo start --all command at system startup to bring up all activated VDO volumes.

You can also create a VDO volume that does not start automatically by adding the --activate=disabled option to the vdo create command.

The starting order

Some systems might place LVM volumes both above VDO volumes and below them. On these systems, it is necessary to start services in the right order:

  1. The lower layer of LVM must start first. In most systems, starting this layer is configured automatically when the LVM package is installed.
  2. The vdo systemd unit must start then.
  3. Finally, additional scripts must run in order to start LVM volumes or other services on top of the running VDO volumes.

How long it takes to stop a volume

Stopping a VDO volume takes time based on the speed of your storage device and the amount of data that the volume needs to write:

  • The volume always writes around 1GiB for every 1GiB of the UDS index.
  • The volume additionally writes the amount of data equal to the block map cache size plus up to 8MiB per slab.
  • The volume must finish processing all outstanding IO requests.

43.2.2.2. Starting a VDO volume

This procedure starts a given VDO volume or all VDO volumes on your system.

Procedure

  • To start a given VDO volume, use:

    # vdo start --name=my-vdo
  • To start all VDO volumes, use:

    # vdo start --all

Additional resources

  • The vdo(8) man page

43.2.2.3. Stopping a VDO volume

This procedure stops a given VDO volume or all VDO volumes on your system.

Procedure

  1. Stop the volume.

    • To stop a given VDO volume, use:

      # vdo stop --name=my-vdo
    • To stop all VDO volumes, use:

      # vdo stop --all
  2. Wait for the volume to finish writing data to the disk.

Additional resources

  • The vdo(8) man page

43.2.3. Automatically starting VDO volumes at system boot

You can configure VDO volumes so that they start automatically at system boot. You can also disable the automatic start.

43.2.3.1. Started and activated VDO volumes

During the system boot, the vdo systemd unit automatically starts all VDO devices that are configured as activated.

The vdo systemd unit is installed and enabled by default when the vdo package is installed. This unit automatically runs the vdo start --all command at system startup to bring up all activated VDO volumes.

You can also create a VDO volume that does not start automatically by adding the --activate=disabled option to the vdo create command.

The starting order

Some systems might place LVM volumes both above VDO volumes and below them. On these systems, it is necessary to start services in the right order:

  1. The lower layer of LVM must start first. In most systems, starting this layer is configured automatically when the LVM package is installed.
  2. The vdo systemd unit must start then.
  3. Finally, additional scripts must run in order to start LVM volumes or other services on top of the running VDO volumes.

How long it takes to stop a volume

Stopping a VDO volume takes time based on the speed of your storage device and the amount of data that the volume needs to write:

  • The volume always writes around 1GiB for every 1GiB of the UDS index.
  • The volume additionally writes the amount of data equal to the block map cache size plus up to 8MiB per slab.
  • The volume must finish processing all outstanding IO requests.

43.2.3.2. Activating a VDO volume

This procedure activates a VDO volume to enable it to start automatically.

Procedure

  • To activate a specific volume:

    # vdo activate --name=my-vdo
  • To activate all volumes:

    # vdo activate --all

Additional resources

  • The vdo(8) man page

43.2.3.3. Deactivating a VDO volume

This procedure deactivates a VDO volume to prevent it from starting automatically.

Procedure

  • To deactivate a specific volume:

    # vdo deactivate --name=my-vdo
  • To deactivate all volumes:

    # vdo deactivate --all

Additional resources

  • The vdo(8) man page

43.2.4. Selecting a VDO write mode

You can configure write mode for a VDO volume, based on what the underlying block device requires. By default, VDO selects write mode automatically.

43.2.4.1. VDO write modes

VDO supports the following write modes:

sync

When VDO is in sync mode, the layers above it assume that a write command writes data to persistent storage. As a result, it is not necessary for the file system or application, for example, to issue FLUSH or force unit access (FUA) requests to cause the data to become persistent at critical points.

VDO must be set to sync mode only when the underlying storage guarantees that data is written to persistent storage when the write command completes. That is, the storage must either have no volatile write cache, or have a write through cache.

async

When VDO is in async mode, VDO does not guarantee that the data is written to persistent storage when a write command is acknowledged. The file system or application must issue FLUSH or FUA requests to ensure data persistence at critical points in each transaction.

VDO must be set to async mode if the underlying storage does not guarantee that data is written to persistent storage when the write command completes; that is, when the storage has a volatile write back cache.

async-unsafe

This mode has the same properties as async but it is not compliant with Atomicity, Consistency, Isolation, Durability (ACID). Compared to async, async-unsafe has a better performance.

Warning

When an application or a file system that assumes ACID compliance operates on top of the VDO volume, async-unsafe mode might cause unexpected data loss.

auto
The auto mode automatically selects sync or async based on the characteristics of each device. This is the default option.

43.2.4.2. The internal processing of VDO write modes

This section provides details on how the sync and async VDO write modes operate.

If the kvdo module is operating in synchronous mode:

  1. It temporarily writes the data in the request to the allocated block and then acknowledges the request.
  2. Once the acknowledgment is complete, an attempt is made to deduplicate the block by computing a MurmurHash-3 signature of the block data, which is sent to the VDO index.
  3. If the VDO index contains an entry for a block with the same signature, kvdo reads the indicated block and does a byte-by-byte comparison of the two blocks to verify that they are identical.
  4. If they are indeed identical, then kvdo updates its block map so that the logical block points to the corresponding physical block and releases the allocated physical block.
  5. If the VDO index did not contain an entry for the signature of the block being written, or the indicated block does not actually contain the same data, kvdo updates its block map to make the temporary physical block permanent.

If kvdo is operating in asynchronous mode:

  1. Instead of writing the data, it will immediately acknowledge the request.
  2. It will then attempt to deduplicate the block in same manner as described above.
  3. If the block turns out to be a duplicate, kvdo updates its block map and releases the allocated block. Otherwise, it writes the data in the request to the allocated block and updates the block map to make the physical block permanent.

43.2.4.3. Checking the write mode on a VDO volume

This procedure lists the active write mode on a selected VDO volume.

Procedure

  • Use the following command to see the write mode used by a VDO volume:

    # vdo status --name=my-vdo

    The output lists:

    • The configured write policy, which is the option selected from sync, async, or auto
    • The write policy, which is the particular write mode that VDO applied, that is either sync or async

43.2.4.4. Checking for a volatile cache

This procedure determines if a block device has a volatile cache or not. You can use the information to choose between the sync and async VDO write modes.

Procedure

  1. Use either of the following methods to determine if a device has a writeback cache:

    • Read the /sys/block/block-device/device/scsi_disk/identifier/cache_type sysfs file. For example:

      $ cat '/sys/block/sda/device/scsi_disk/7:0:0:0/cache_type'
      
      write back
      $ cat '/sys/block/sdb/device/scsi_disk/1:2:0:0/cache_type'
      
      None
    • Alternatively, you can find whether the above mentioned devices have a write cache or not in the kernel boot log:

      sd 7:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
      sd 1:2:0:0: [sdb] Write cache: disabled, read cache: disabled, supports DPO and FUA
  2. In the previous examples:

    • Device sda indicates that it has a writeback cache. Use async mode for it.
    • Device sdb indicates that it does not have a writeback cache. Use sync mode for it.

    You should configure VDO to use the sync write mode if the cache_type value is None or write through.

43.2.4.5. Setting a VDO write mode

This procedure sets a write mode for a VDO volume, either for an existing one or when creating a new volume.

Important

Using an incorrect write mode might result in data loss after a power failure, a system crash, or any unexpected loss of contact with the disk.

Prerequisites

Procedure

  • You can set a write mode either on an existing VDO volume or when creating a new volume:

    • To modify an existing VDO volume, use:

      # vdo changeWritePolicy --writePolicy=sync|async|async-unsafe|auto \
                              --name=vdo-name
    • To specify a write mode when creating a VDO volume, add the --writePolicy=sync|async|async-unsafe|auto option to the vdo create command.

43.2.5. Recovering a VDO volume after an unclean shutdown

You can recover a VDO volume after an unclean shutdown to enable it to continue operating. The task is mostly automated. Additionally, you can clean up after a VDO volume was unsuccessfully created because of a failure in the process.

43.2.5.1. VDO write modes

VDO supports the following write modes:

sync

When VDO is in sync mode, the layers above it assume that a write command writes data to persistent storage. As a result, it is not necessary for the file system or application, for example, to issue FLUSH or force unit access (FUA) requests to cause the data to become persistent at critical points.

VDO must be set to sync mode only when the underlying storage guarantees that data is written to persistent storage when the write command completes. That is, the storage must either have no volatile write cache, or have a write through cache.

async

When VDO is in async mode, VDO does not guarantee that the data is written to persistent storage when a write command is acknowledged. The file system or application must issue FLUSH or FUA requests to ensure data persistence at critical points in each transaction.

VDO must be set to async mode if the underlying storage does not guarantee that data is written to persistent storage when the write command completes; that is, when the storage has a volatile write back cache.

async-unsafe

This mode has the same properties as async but it is not compliant with Atomicity, Consistency, Isolation, Durability (ACID). Compared to async, async-unsafe has a better performance.

Warning

When an application or a file system that assumes ACID compliance operates on top of the VDO volume, async-unsafe mode might cause unexpected data loss.

auto
The auto mode automatically selects sync or async based on the characteristics of each device. This is the default option.

43.2.5.2. VDO volume recovery

When a VDO volume restarts after an unclean shutdown, VDO performs the following actions:

  • Verifies the consistency of the metadata on the volume.
  • Rebuilds a portion of the metadata to repair it if necessary.

Rebuilds are automatic and do not require user intervention.

VDO might rebuild different writes depending on the active write mode:

sync
If VDO was running on synchronous storage and write policy was set to sync, all data written to the volume are fully recovered.
async
If the write policy was async, some writes might not be recovered if they were not made durable. This is done by sending VDO a FLUSH command or a write I/O tagged with the FUA (force unit access) flag. You can accomplish this from user mode by invoking a data integrity operation like fsync, fdatasync, sync, or umount.

In either mode, some writes that were either unacknowledged or not followed by a flush might also be rebuilt.

Automatic and manual recovery

When a VDO volume enters recovering operating mode, VDO automatically rebuilds the unclean VDO volume after the it comes back online. This is called online recovery.

If VDO cannot recover a VDO volume successfully, it places the volume in read-only operating mode that persists across volume restarts. You need to fix the problem manually by forcing a rebuild.

Additional resources

43.2.5.3. VDO operating modes

This section describes the modes that indicate whether a VDO volume is operating normally or is recovering from an error.

You can display the current operating mode of a VDO volume using the vdostats --verbose device command. See the Operating mode attribute in the output.

normal
This is the default operating mode. VDO volumes are always in normal mode, unless either of the following states forces a different mode. A newly created VDO volume starts in normal mode.
recovering

When a VDO volume does not save all of its metadata before shutting down, it automatically enters recovering mode the next time that it starts up. The typical reasons for entering this mode are sudden power loss or a problem from the underlying storage device.

In recovering mode, VDO is fixing the references counts for each physical block of data on the device. Recovery usually does not take very long. The time depends on how large the VDO volume is, how fast the underlying storage device is, and how many other requests VDO is handling simultaneously. The VDO volume functions normally with the following exceptions:

  • Initially, the amount of space available for write requests on the volume might be limited. As more of the metadata is recovered, more free space becomes available.
  • Data written while the VDO volume is recovering might fail to deduplicate against data written before the crash if that data is in a portion of the volume that has not yet been recovered. VDO can compress data while recovering the volume. You can still read or overwrite compressed blocks.
  • During an online recovery, certain statistics are unavailable: for example, blocks in use and blocks free. These statistics become available when the rebuild is complete.
  • Response times for reads and writes might be slower than usual due to the ongoing recovery work

You can safely shut down the VDO volume in recovering mode. If the recovery does not finish before shutting down, the device enters recovering mode again the next time that it starts up.

The VDO volume automatically exits recovering mode and moves to normal mode when it has fixed all the reference counts. No administrator action is necessary. For details, see Section 43.2.5.4, “Recovering a VDO volume online”.

read-only

When a VDO volume encounters a fatal internal error, it enters read-only mode. Events that might cause read-only mode include metadata corruption or the backing storage device becoming read-only. This mode is an error state.

In read-only mode, data reads work normally but data writes always fail. The VDO volume stays in read-only mode until an administrator fixes the problem.

You can safely shut down a VDO volume in read-only mode. The mode usually persists after the VDO volume is restarted. In rare cases, the VDO volume is not able to record the read-only state to the backing storage device. In these cases, VDO attempts to do a recovery instead.

Once a volume is in read-only mode, there is no guarantee that data on the volume has not been lost or corrupted. In such cases, Red Hat recommends copying the data out of the read-only volume and possibly restoring the volume from backup.

If the risk of data corruption is acceptable, it is possible to force an offline rebuild of the VDO volume metadata so the volume can be brought back online and made available. The integrity of the rebuilt data cannot be guaranteed. For details, see Section 43.2.5.5, “Forcing an offline rebuild of a VDO volume metadata”.

43.2.5.4. Recovering a VDO volume online

This procedure performs an online recovery on a VDO volume to recover metadata after an unclean shutdown.

Procedure

  1. If the VDO volume is not already started, start it:

    # vdo start --name=my-vdo

    No additional steps are necessary. The recovery runs in the background.

  2. If you rely on volume statistics like blocks in use and blocks free, wait until they are available.

43.2.5.5. Forcing an offline rebuild of a VDO volume metadata

This procedure performs a forced offline rebuild of a VDO volume metadata to recover after an unclean shutdown.

Warning

This procedure might cause data loss on the volume.

Prerequisites

  • The VDO volume is started.

Procedure

  1. Check if the volume is in read-only mode. See the operating mode attribute in the command output:

    # vdo status --name=my-vdo

    If the volume is not in read-only mode, it is not necessary to force an offline rebuild. Perform an online recovery as described in Section 43.2.5.4, “Recovering a VDO volume online”.

  2. Stop the volume if it is running:

    # vdo stop --name=my-vdo
  3. Restart the volume with the --forceRebuild option:

    # vdo start --name=my-vdo --forceRebuild

43.2.5.6. Removing an unsuccessfully created VDO volume

This procedure cleans up a VDO volume in an intermediate state. A volume is left in an intermediate state if a failure occurs when creating the volume. This might happen when, for example:

  • The system crashes
  • Power fails
  • The administrator interrupts a running vdo create command

Procedure

  • To clean up, remove the unsuccessfully created volume with the --force option:

    # vdo remove --force --name=my-vdo

    The --force option is required because the administrator might have caused a conflict by changing the system configuration since the volume was unsuccessfully created.

    Without the --force option, the vdo remove command fails with the following message:

    [...]
    A previous operation failed.
    Recovery from the failure either failed or was interrupted.
    Add '--force' to 'remove' to perform the following cleanup.
    Steps to clean up VDO my-vdo:
    umount -f /dev/mapper/my-vdo
    udevadm settle
    dmsetup remove my-vdo
    vdo: ERROR - VDO volume my-vdo previous operation (create) is incomplete

43.2.6. Optimizing the UDS index

You can configure certain settings of the UDS index to optimize it on your system.

Important

You cannot change the properties of the UDS index after creating the VDO volume.

43.2.6.1. Components of a VDO volume

VDO uses a block device as a backing store, which can include an aggregation of physical storage consisting of one or more disks, partitions, or even flat files. When a storage management tool creates a VDO volume, VDO reserves volume space for the UDS index and VDO volume. The UDS index and the VDO volume interact together to provide deduplicated block storage.

Figure 43.4. VDO disk organization

VDO disk organization

The VDO solution consists of the following components:

kvdo

A kernel module that loads into the Linux Device Mapper layer provides a deduplicated, compressed, and thinly provisioned block storage volume.

The kvdo module exposes a block device. You can access this block device directly for block storage or present it through a Linux file system, such as XFS or ext4.

When kvdo receives a request to read a logical block of data from a VDO volume, it maps the requested logical block to the underlying physical block and then reads and returns the requested data.

When kvdo receives a request to write a block of data to a VDO volume, it first checks whether the request is a DISCARD or TRIM request or whether the data is uniformly zero. If either of these conditions is true, kvdo updates its block map and acknowledges the request. Otherwise, VDO processes and optimizes the data.

uds

A kernel module that communicates with the Universal Deduplication Service (UDS) index on the volume and analyzes data for duplicates. For each new piece of data, UDS quickly determines if that piece is identical to any previously stored piece of data. If the index finds a match, the storage system can then internally reference the existing item to avoid storing the same information more than once.

The UDS index runs inside the kernel as the uds kernel module.

Command line tools
For configuring and managing optimized storage.

43.2.6.2. The UDS index

VDO uses a high-performance deduplication index called UDS to detect duplicate blocks of data as they are being stored.

The UDS index provides the foundation of the VDO product. For each new piece of data, it quickly determines if that piece is identical to any previously stored piece of data. If the index finds match, the storage system can then internally reference the existing item to avoid storing the same information more than once.

The UDS index runs inside the kernel as the uds kernel module.

The deduplication window is the number of previously written blocks that the index remembers. The size of the deduplication window is configurable. For a given window size, the index requires a specific amount of RAM and a specific amount of disk space. The size of the window is usually determined by specifying the size of the index memory using the --indexMem=size option. VDO then determines the amount of disk space to use automatically.

The UDS index consists of two parts:

  • A compact representation is used in memory that contains at most one entry per unique block.
  • An on-disk component that records the associated block names presented to the index as they occur, in order.

UDS uses an average of 4 bytes per entry in memory, including cache.

The on-disk component maintains a bounded history of data passed to UDS. UDS provides deduplication advice for data that falls within this deduplication window, containing the names of the most recently seen blocks. The deduplication window allows UDS to index data as efficiently as possible while limiting the amount of memory required to index large data repositories. Despite the bounded nature of the deduplication window, most datasets which have high levels of deduplication also exhibit a high degree of temporal locality — in other words, most deduplication occurs among sets of blocks that were written at about the same time. Furthermore, in general, data being written is more likely to duplicate data that was recently written than data that was written a long time ago. Therefore, for a given workload over a given time interval, deduplication rates will often be the same whether UDS indexes only the most recent data or all the data.

Because duplicate data tends to exhibit temporal locality, it is rarely necessary to index every block in the storage system. Were this not so, the cost of index memory would outstrip the savings of reduced storage costs from deduplication. Index size requirements are more closely related to the rate of data ingestion. For example, consider a storage system with 100 TB of total capacity but with an ingestion rate of 1 TB per week. With a deduplication window of 4 TB, UDS can detect most redundancy among the data written within the last month.

43.2.7. Enabling or disabling deduplication in VDO

In some instances, you might want to temporarily disable deduplication of data being written to a VDO volume while still retaining the ability to read to and write from the volume. Disabling deduplication prevents subsequent writes from being deduplicated, but the data that was already deduplicated remains so.

43.2.7.1. Deduplication in VDO

Deduplication is a technique for reducing the consumption of storage resources by eliminating multiple copies of duplicate blocks.

Instead of writing the same data more than once, VDO detects each duplicate block and records it as a reference to the original block. VDO maintains a mapping from logical block addresses, which are used by the storage layer above VDO, to physical block addresses, which are used by the storage layer under VDO.

After deduplication, multiple logical block addresses can be mapped to the same physical block address. These are called shared blocks. Block sharing is invisible to users of the storage, who read and write blocks as they would if VDO were not present.

When a shared block is overwritten, VDO allocates a new physical block for storing the new block data to ensure that other logical block addresses that are mapped to the shared physical block are not modified.

43.2.7.2. Enabling deduplication on a VDO volume

This procedure restarts the associated UDS index and informs the VDO volume that deduplication is active again.

Note

Deduplication is enabled by default.

Procedure

  • To restart deduplication on a VDO volume, use the following command:

    # vdo enableDeduplication --name=my-vdo

43.2.7.3. Disabling deduplication on a VDO volume

This procedure stops the associated UDS index and informs the VDO volume that deduplication is no longer active.

Procedure

  • To stop deduplication on a VDO volume, use the following command:

    # vdo disableDeduplication --name=my-vdo
  • You can also disable deduplication when creating a new VDO volume by adding the --deduplication=disabled option to the vdo create command.

43.2.8. Enabling or disabling compression in VDO

VDO provides data compression. You can disable it to maximize performance or to speed processing of data that is unlikely to compress, or re-enable it to increase space savings.

43.2.8.1. Compression in VDO

In addition to block-level deduplication, VDO also provides inline block-level compression using the HIOPS Compression™ technology.

VDO volume compression is on by default.

While deduplication is the optimal solution for virtual machine environments and backup applications, compression works very well with structured and unstructured file formats that do not typically exhibit block-level redundancy, such as log files and databases.

Compression operates on blocks that have not been identified as duplicates. When VDO sees unique data for the first time, it compresses the data. Subsequent copies of data that have already been stored are deduplicated without requiring an additional compression step.

The compression feature is based on a parallelized packaging algorithm that enables it to handle many compression operations at once. After first storing the block and responding to the requestor, a best-fit packing algorithm finds multiple blocks that, when compressed, can fit into a single physical block. After it is determined that a particular physical block is unlikely to hold additional compressed blocks, it is written to storage and the uncompressed blocks are freed and reused.

By performing the compression and packaging operations after having already responded to the requestor, using compression imposes a minimal latency penalty.

43.2.8.2. Enabling compression on a VDO volume

This procedure enables compression on a VDO volume to increase space savings.

Note

Compression is enabled by default.

Procedure

  • To start it again, use the following command:

    # vdo enableCompression --name=my-vdo

43.2.8.3. Disabling compression on a VDO volume

This procedure stops compression on a VDO volume to maximize performance or to speed processing of data that is unlikely to compress.

Procedure

  • To stop compression on an existing VDO volume, use the following command:

    # vdo disableCompression --name=my-vdo
  • Alternatively, you can disable compression by adding the --compression=disabled option to the vdo create command when creating a new volume.

43.2.9. Increasing the size of a VDO volume

You can increase the physical size of a VDO volume to utilize more underlying storage capacity, or the logical size to provide more capacity on the volume.

43.2.9.1. The physical and logical size of a VDO volume

This section describes the physical size, available physical size, and logical size that VDO can utilize:

Physical size

This is the same size as the underlying block device. VDO uses this storage for:

  • User data, which might be deduplicated and compressed
  • VDO metadata, such as the UDS index
Available physical size

This is the portion of the physical size that VDO is able to use for user data

It is equivalent to the physical size minus the size of the metadata, minus the remainder after dividing the volume into slabs by the given slab size.

Logical Size

This is the provisioned size that the VDO volume presents to applications. It is usually larger than the available physical size. If the --vdoLogicalSize option is not specified, then the provisioning of the logical volume is now provisioned to a 1:1 ratio. For example, if a VDO volume is put on top of a 20 GB block device, then 2.5 GB is reserved for the UDS index (if the default index size is used). The remaining 17.5 GB is provided for the VDO metadata and user data. As a result, the available storage to consume is not more than 17.5 GB, and can be less due to metadata that makes up the actual VDO volume.

VDO currently supports any logical size up to 254 times the size of the physical volume with an absolute maximum logical size of 4PB.

Figure 43.5. VDO disk organization

VDO disk organization

In this figure, the VDO deduplicated storage target sits completely on top of the block device, meaning the physical size of the VDO volume is the same size as the underlying block device.

Additional resources

43.2.9.2. Thin provisioning in VDO

VDO is a thinly provisioned block storage target. The amount of physical space that a VDO volume uses might differ from the size of the volume that is presented to users of the storage. You can make use of this disparity to save on storage costs.

Out-of-space conditions

Take care to avoid unexpectedly running out of storage space, if the data written does not achieve the expected rate of optimization.

Whenever the number of logical blocks (virtual storage) exceeds the number of physical blocks (actual storage), it becomes possible for file systems and applications to unexpectedly run out of space. For that reason, storage systems using VDO must provide you with a way of monitoring the size of the free pool on the VDO volume.

You can determine the size of this free pool by using the vdostats utility. The default output of this utility lists information for all running VDO volumes in a format similar to the Linux df utility. For example:

Device                1K-blocks   Used        Available   Use%
/dev/mapper/vdo-name  211812352   105906176   105906176   50%

When the physical storage capacity of a VDO volume is almost full, VDO reports a warning in the system log, similar to the following:

Oct  2 17:13:39 system lvm[13863]: Monitoring VDO pool vdo-name.
Oct  2 17:27:39 system lvm[13863]: WARNING: VDO pool vdo-name is now 80.69% full.
Oct  2 17:28:19 system lvm[13863]: WARNING: VDO pool vdo-name is now 85.25% full.
Oct  2 17:29:39 system lvm[13863]: WARNING: VDO pool vdo-name is now 90.64% full.
Oct  2 17:30:29 system lvm[13863]: WARNING: VDO pool vdo-name is now 96.07% full.
Note

These warning messages appear only when the lvm2-monitor service is running. It is enabled by default.

How to prevent out-of-space conditions

If the size of free pool drops below a certain level, you can take action by:

  • Deleting data. This reclaims space whenever the deleted data is not duplicated. Deleting data frees the space only after discards are issued.
  • Adding physical storage
Important

Monitor physical space on your VDO volumes to prevent out-of-space situations. Running out of physical blocks might result in losing recently written, unacknowledged data on the VDO volume.

Thin provisioning and the TRIM and DISCARD commands

To benefit from the storage savings of thin provisioning, the physical storage layer needs to know when data is deleted. File systems that work with thinly provisioned storage send TRIM or DISCARD commands to inform the storage system when a logical block is no longer required.

Several methods of sending the TRIM or DISCARD commands are available:

  • With the discard mount option, the file systems can send these commands whenever a block is deleted.
  • You can send the commands in a controlled manner by using utilities such as fstrim. These utilities tell the file system to detect which logical blocks are unused and send the information to the storage system in the form of a TRIM or DISCARD command.

The need to use TRIM or DISCARD on unused blocks is not unique to VDO. Any thinly provisioned storage system has the same challenge.

43.2.9.3. Increasing the logical size of a VDO volume

This procedure increases the logical size of a given VDO volume. It enables you to initially create VDO volumes that have a logical size small enough to be safe from running out of space. After some period of time, you can evaluate the actual rate of data reduction, and if sufficient, you can grow the logical size of the VDO volume to take advantage of the space savings.

It is not possible to decrease the logical size of a VDO volume.

Procedure

  • To grow the logical size, use:

    # vdo growLogical --name=my-vdo \
                      --vdoLogicalSize=new-logical-size

    When the logical size increases, VDO informs any devices or file systems on top of the volume of the new size.

43.2.9.4. Increasing the physical size of a VDO volume

This procedure increases the amount of physical storage available to a VDO volume.

It is not possible to shrink a VDO volume in this way.

Prerequisites

  • The underlying block device has a larger capacity than the current physical size of the VDO volume.

    If it does not, you can attempt to increase the size of the device. The exact procedure depends on the type of the device. For example, to resize an MBR or GPT partition, see the Resizing a partition section in the Managing storage devices guide.

Procedure

  • Add the new physical storage space to the VDO volume:

    # vdo growPhysical --name=my-vdo

43.2.10. Removing VDO volumes

You can remove an existing VDO volume on your system.

43.2.10.1. Removing a working VDO volume

This procedure removes a VDO volume and its associated UDS index.

Procedure

  1. Unmount the file systems and stop the applications that are using the storage on the VDO volume.
  2. To remove the VDO volume from your system, use:

    # vdo remove --name=my-vdo

43.2.10.2. Removing an unsuccessfully created VDO volume

This procedure cleans up a VDO volume in an intermediate state. A volume is left in an intermediate state if a failure occurs when creating the volume. This might happen when, for example:

  • The system crashes
  • Power fails
  • The administrator interrupts a running vdo create command

Procedure

  • To clean up, remove the unsuccessfully created volume with the --force option:

    # vdo remove --force --name=my-vdo

    The --force option is required because the administrator might have caused a conflict by changing the system configuration since the volume was unsuccessfully created.

    Without the --force option, the vdo remove command fails with the following message:

    [...]
    A previous operation failed.
    Recovery from the failure either failed or was interrupted.
    Add '--force' to 'remove' to perform the following cleanup.
    Steps to clean up VDO my-vdo:
    umount -f /dev/mapper/my-vdo
    udevadm settle
    dmsetup remove my-vdo
    vdo: ERROR - VDO volume my-vdo previous operation (create) is incomplete

43.3. Discarding unused blocks

You can perform or schedule discard operations on block devices that support them.

43.3.1. Block discard operations

Block discard operations discard blocks that are no longer in use by a mounted file system. They are useful on:

  • Solid-state drives (SSDs)
  • Thinly-provisioned storage
Requirements

The block device underlying the file system must support physical discard operations.

Physical discard operations are supported if the value in the /sys/block/device/queue/discard_max_bytes file is not zero.

43.3.2. Types of block discard operations

You can run discard operations using different methods:

Batch discard
Are run explicitly by the user. They discard all unused blocks in the selected file systems.
Online discard
Are specified at mount time. They run in real time without user intervention. Online discard operations discard only the blocks that are transitioning from used to free.
Periodic discard
Are batch operations that are run regularly by a systemd service.

All types are supported by the XFS and ext4 file systems and by VDO.

Recommendations

Red Hat recommends that you use batch or periodic discard.

Use online discard only if:

  • the system’s workload is such that batch discard is not feasible, or
  • online discard operations are necessary to maintain performance.

43.3.3. Performing batch block discard

This procedure performs a batch block discard operation to discard unused blocks on a mounted file system.

Prerequisites

  • The file system is mounted.
  • The block device underlying the file system supports physical discard operations.

Procedure

  • Use the fstrim utility:

    • To perform discard only on a selected file system, use:

      # fstrim mount-point
    • To perform discard on all mounted file systems, use:

      # fstrim --all

If you execute the fstrim command on:

  • a device that does not support discard operations, or
  • a logical device (LVM or MD) composed of multiple devices, where any one of the device does not support discard operations,

the following message displays:

# fstrim /mnt/non_discard

fstrim: /mnt/non_discard: the discard operation is not supported

Additional resources

  • The fstrim(8) man page

43.3.4. Enabling online block discard

This procedure enables online block discard operations that automatically discard unused blocks on all supported file systems.

Procedure

  • Enable online discard at mount time:

    • When mounting a file system manually, add the -o discard mount option:

      # mount -o discard device mount-point
    • When mounting a file system persistently, add the discard option to the mount entry in the /etc/fstab file.

Additional resources

  • The mount(8) man page
  • The fstab(5) man page

43.3.5. Enabling online block discard using RHEL System Roles

This section describes how to enable online block discard using the storage role.

Prerequisites

  • An Ansible playbook including the storage role exists.

For information on how to apply such a playbook, see Applying a role.

43.3.5.1. Example Ansible playbook to enable online block discard

This section provides an example Ansible playbook. This playbook applies the storage role to mount an XFS file system with online block discard enabled.

Example 43.2. A playbook that enables online block discard on /mnt/data/

---
- hosts: all
  vars:
    storage_volumes:
      - name: barefs
        type: disk
        disks:
          - sdb
        fs_type: xfs
        mount_point: /mnt/data
        mount_options: discard
  roles:
    - rhel-system-roles.storage

Additional resources

  • For details about the parameters used in the storage system role, see the /usr/share/ansible/roles/rhel-system-roles.storage/README.md file.

43.3.6. Enabling periodic block discard

This procedure enables a systemd timer that regularly discards unused blocks on all supported file systems.

Procedure

  • Enable and start the systemd timer:

    # systemctl enable --now fstrim.timer

43.4. Using the web console for managing Virtual Data Optimizer volumes

Configure the Virtual Data Optimizer (VDO) using the RHEL 8 web console.

You will learn how to:

  • Create VDO volumes
  • Format VDO volumes
  • Extend VDO volumes

Prerequisites

  • The RHEL 8 web console is installed and accessible.

    For details, see Installing the web console.

  • The cockpit-storaged package is installed on your system.

43.4.1. VDO volumes in the web console

Red Hat Enterprise Linux 8 supports Virtual Data Optimizer (VDO).

VDO is a block virtualization technology that combines:

Compression
For details, see Enabling or disabling compression in VDO.
Deduplication
For details, see Enabling or disabling deduplication in VDO.
Thin provisioning
For details, see Thinly-provisioned logical volumes (thin volumes).

Using these technologies, VDO:

  • Saves storage space inline
  • Compresses files
  • Eliminates duplications
  • Enables you to allocate more virtual space than how much the physical or logical storage provides
  • Enables you to extend the virtual storage by growing

VDO can be created on top of many types of storage. In the RHEL 8 web console, you can configure VDO on top of:

  • LVM

    Note

    It is not possible to configure VDO on top of thinly-provisioned volumes.

  • Physical volume
  • Software RAID

For details about placement of VDO in the Storage Stack, see System Requirements.

Additional resources

43.4.2. Creating VDO volumes in the web console

Create a VDO volume in the RHEL web console.

Prerequisites

  • Physical drives, LVMs, or RAID from which you want to create VDO.

Procedure

  1. Log in to the RHEL 8 web console.

    For details, see Logging in to the web console.

  2. Click Storage.
  3. Click the + icon in the VDO Devices box.

    cockpit adding vdo

  4. In the Name field, enter a name of a VDO volume without spaces.
  5. Select the drive that you want to use.
  6. In the Logical Size bar, set up the size of the VDO volume. You can extend it more than ten times, but consider for what purpose you are creating the VDO volume:

    • For active VMs or container storage, use logical size that is ten times the physical size of the volume.
    • For object storage, use logical size that is three times the physical size of the volume.

    For details, see Deploying VDO.

  7. In the Index Memory bar, allocate memory for the VDO volume.

    For details about VDO system requirements, see System Requirements.

  8. Select the Compression option. This option can efficiently reduce various file formats.

    For details, see Enabling or disabling compression in VDO.

  9. Select the Deduplication option.

    This option reduces the consumption of storage resources by eliminating multiple copies of duplicate blocks. For details, see Enabling or disabling deduplication in VDO.

  10. [Optional] If you want to use the VDO volume with applications that need a 512 bytes block size, select Use 512 Byte emulation. This reduces the performance of the VDO volume, but should be very rarely needed. If in doubt, leave it off.
  11. Click Create.

    cockpit create vdo dialog

If the process of creating the VDO volume succeeds, you can see the new VDO volume in the Storage section and format it with a file system.

cockpit vdo created

43.4.3. Formatting VDO volumes in the web console

VDO volumes act as physical drives. To use them, you need to format them with a file system.

Warning

Formatting VDO will erase all data on the volume.

The following steps describe the procedure to format VDO volumes.

Prerequisites

Procedure

  1. Log in to the RHEL 8 web console.

    For details, see Logging in to the web console.

  2. Click Storage.
  3. Click the VDO volume.
  4. Click on the Unrecognized Data tab.
  5. Click Format.

    cockpit vdo format

  6. In the Erase drop down menu, select:

    Don’t overwrite existing data
    The RHEL web console rewrites only the disk header. The advantage of this option is the speed of formatting.
    Overwrite existing data with zeros
    The RHEL web console rewrites the whole disk with zeros. This option is slower because the program has to go through the whole disk. Use this option if the disk includes any data and you need to rewrite them.
  7. In the Type drop down menu, select a filesystem:

    • The XFS file system supports large logical volumes, switching physical drives online without outage, and growing. Leave this file system selected if you do not have a different strong preference.

      XFS does not support shrinking volumes. Therefore, you will not be able to reduce volume formatted with XFS.

    • The ext4 file system supports logical volumes, switching physical drives online without outage, growing, and shrinking.

    You can also select a version with the LUKS (Linux Unified Key Setup) encryption, which allows you to encrypt the volume with a passphrase.

  8. In the Name field, enter the logical volume name.
  9. In the Mounting drop down menu, select Custom.

    The Default option does not ensure that the file system will be mounted on the next boot.

  10. In the Mount Point field, add the mount path.
  11. Select Mount at boot.

    cockpit lv format

  12. Click Format.

    Formatting can take several minutes depending on the used formatting options and the volume size.

    After a successful finish, you can see the details of the formatted VDO volume on the Filesystem tab.

    cockpit vdo formatted

  13. To use the VDO volume, click Mount.

At this point, the system uses the mounted and formatted VDO volume.

43.4.4. Extending VDO volumes in the web console

Extend VDO volumes in the RHEL 8 web console.

Prerequisites

  • The cockpit-storaged package is installed on your system.
  • The VDO volume created.

Procedure

  1. Log in to the RHEL 8 web console.

    For details, see Logging in to the web console.

  2. Click Storage.
  3. Click your VDO volume in the VDO Devices box.

    cockpit vdo created

  4. In the VDO volume details, click the Grow button.
  5. In the Grow logical size of VDO dialog box, extend the logical size of the VDO volume.

    cockpit vdo grow done

    Original size of the logical volume from the screenshot was 6 GB. As you can see, the RHEL web console enables you to grow the volume to more than ten times the size and it works correctly because of the compression and deduplication.

  6. Click Grow.

If the process of growing VDO succeeds, you can see the new size in the VDO volume details.

cockpit vdo grow details