8. Storage

The Storage section of the graphical interface allows configuration of these options:

8.1. Volumes

The Volumes section of the FreeNAS® graphical interface can be used to format ZFS pools, import a disk to copy its data into an existing pool, or import an existing ZFS pool. It can also be used to create ZFS datasets and zvols and to manage their permissions.

Note

In ZFS terminology, the storage that is managed by ZFS is referred to as a pool. The FreeNAS® graphical interface uses the term volume to refer to a ZFS pool.

Proper storage design is important for any NAS. Please read through this entire chapter before configuring storage disks. All of the features are described to help make it clear which will be the most benefit for your uses, and caveats or caveats or hardware restrictions which could limit their use.

8.1.1. Volume Manager

Volume Manager is used to add disks to a ZFS pool. Any old data on added disks is overwritten, so save it elsewhere before reusing a disk. Please see the ZFS Primer for information on ZFS redundancy with multiple disks before using Volume Manager.

Selecting Storage Volumes Volume Manager opens a screen like the example shown in Figure 8.1.1.

_images/zfs1a.png

Fig. 8.1.1 Creating a ZFS Pool Using Volume Manager

Table 8.1.1 summarizes the configuration options of this screen.

Table 8.1.1 Options When Creating a ZFS Volume
Setting Value Description
Volume name string ZFS volumes must conform to these naming conventions; it is recommended to choose a name that will stick out in the logs (e.g. not data or freenas)
Volume to extend drop-down menu used to extend an existing ZFS pool; see Extending a ZFS Volume for instructions
Encryption checkbox read the section on Encryption before choosing to use encryption
Available disks display displays the number and size of available disks; hover over show to list the available device names; click the + to add all of the disks to the pool
Volume layout drag and drop click and drag the icon to select the desired number of disks for a vdev; once at least one disk is selected, the layouts supported by the selected number of disks will be added to the drop-down menu
Add Extra Device button used to configure multiple vdevs or to add log or cache devices during pool creation
Manual setup button used to create a pool manually (not recommended); see Manual Setup for details

Drag the slider to select the desired number of disks. Volume Manager displays the resulting storage capacity, including taking swap space into account. To change the layout or the number of disks, use the mouse to drag the slider to the desired volume layout. The Volume layout drop-down menu can also be clicked if a different level of redundancy is required.

Note

For performance and capacity reasons, this screen does not allow creating a volume from disks of differing sizes. While it is not recommended, it is possible to create a volume in this situation by using the Manual setup button and following the instructions in Manual Setup.

Volume Manager only allows choosing a configuration if enough disks have been selected to create that configuration. These layouts are supported:

  • Stripe: requires at least one disk
  • Mirror: requires at least two disks
  • RAIDZ1: requires at least three disks
  • RAIDZ2: requires at least four disks
  • RAIDZ3: requires at least five disks
  • log device: requires at least one dedicated device, a fast, low-latency, power-protected SSD is recommended
  • cache device: requires at least one dedicated device, SSD is recommended

When more than five disks are used, consideration must be given to the optimal layout for the best performance and scalability. An overview of the recommended disk group sizes as well as more information about log and cache devices can be found in the ZFS Primer.

The Add Volume button warns that existing data will be cleared. In other words, creating a new volume reformats the selected disks. If the existing data is meant to be preserved, click the Cancel button and refer to Import Disk and Import Volume to see if the existing format is supported. If so, perform that action instead. If the current storage format is not supported, it is necessary to back up the data to external media, format the disks, then restore the data to the new volume.

Depending on the size and number of disks, the type of controller, and whether encryption is selected, creating the volume may take some time. After the volume is created, the screen will refresh and the new volume is listed in the tree under Storage Volumes. Click the + next to the volume name to access its Change Permissions, Create Dataset, and Create zvol options.

8.1.1.1. Encryption

FreeNAS® supports GELI full disk encryption for ZFS volumes. It is important to understand the details when considering whether encryption is right for your FreeNAS® system:

  • This is not the encryption method used by Oracle’s version of ZFS. That version is not open source and is the property of Oracle.
  • This is full disk encryption and not per-filesystem encryption. The underlying drives are first encrypted, then the pool is created on top of the encrypted devices.
  • This type of encryption is primarily targeted at users who store sensitive data and want to retain the ability to remove disks from the pool without having to first wipe the disk’s contents.
  • This design is only suitable for safe disposal of disks independent of the encryption key. As long as the key and the disks are intact, the system is vulnerable to being decrypted. The key should be protected by a strong passphrase and any backups of the key should be securely stored.
  • On the other hand, if the key is lost, the data on the disks is inaccessible. Always back up the key!
  • The encryption key is per ZFS volume (pool). Multiple pools each have their own encryption key.
  • If the system has a lot of disks, performance will suffer if the CPU does not support AES-NI or if no crypto hardware is installed. Without hardware acceleration, there will be about a 20% performance decrease for a single disk. Performance degradation increases with more disks. As data is written, it is automatically encrypted. As data is read, it is decrypted on the fly. If the processor supports the AES-NI instruction set, there is very little, if any, degradation in performance when using encryption. This forum post compares the performance of various CPUs.
  • Data in the ARC cache and the contents of RAM are unencrypted.
  • Swap is always encrypted, even on unencrypted volumes.
  • There is no way to convert an existing, unencrypted volume. Instead, the data must be backed up, the existing pool destroyed, a new encrypted volume created, and the backup restored to the new volume.
  • Hybrid pools are not supported. In other words, newly created vdevs must match the existing encryption scheme. When extending a volume, Volume Manager automatically encrypts the new vdev being added to the existing encrypted pool.

Note

The encryption facility used by FreeNAS® is designed to protect against physical theft of the disks. It is not designed to protect against unauthorized software access. Ensure that only authorized users have access to the administrative GUI and that proper permissions are set on shares if sensitive data is stored on the system.

To create an encrypted volume, check the Encryption box shown in Figure 8.1.1. A pop-up message shows a reminder that it is extremely important to make a backup of the key. Without the key, the data on the disks is inaccessible. Refer to Managing Encrypted Volumes for instructions.

8.1.1.2. Manual Setup

The Manual Setup button shown in Figure 8.1.1 can be used to create a ZFS volume manually. While this is not recommended, it can, for example, be used to create a non-optimal volume containing disks of different sizes.

Note

The usable space of each disk in a volume is limited to the size of the smallest disk in the volume. Because of this, creating volumes with disks of the same size through the Volume Manager is recommended.

Figure 8.1.2 shows the Manual Setup screen. Table 8.1.2 shows the available options.

_images/manual.png

Fig. 8.1.2 Manually Creating a ZFS Volume

Note

Because of the disadvantages of creating volumes with disks of different sizes, the displayed list of disks is sorted by size.

Table 8.1.2 Manual Setup Options
Setting Value Description
Volume name string ZFS volumes must conform to these naming conventions ; it is recommended to choose a name that will stick out in the logs (e.g. not data or freenas)
Encryption checkbox read the section on Encryption before choosing to use encryption
Member disks list highlight desired number of disks from list of available disks
Deduplication drop-down menu choices are Off, Verify, and On; carefully consider the section on Deduplication before changing this setting
ZFS Extra bullet selection used to specify if disk is used for storage (None), a log device, a cache device, or a spare

8.1.1.3. Extending a ZFS Volume

The Volume to extend drop-down menu in Storage Volumes Volume Manager, shown in Figure 8.1.1, can be used to add additional disks to an existing ZFS volume. This drop-down menu will be empty if no ZFS volume exists.

Note

If the existing volume is encrypted, a warning message will remind you that the operation of extending a volume will reset the passphrase and recovery key. After extending the volume, you should immediately recreate both using the instructions in Managing Encrypted Volumes.

After an existing volume has been selected from the drop-down menu, drag and drop the desired disks and select the desired volume layout. For example, disks can be added to increase the capacity of the ZFS pool.

When adding disks to increase the capacity of a volume, ZFS supports the addition of virtual devices, known as vdevs, to an existing ZFS pool. A vdev can be a single disk, a stripe, a mirror, a RAIDZ1, RAIDZ2, or a RAIDZ3. After a vdev is created, more drives cannot be added to that vdev; however, you can stripe a new vdev (and its disks) with another of the same type of existing vdev to increase the overall size of ZFS the pool. In other words, when you extend a ZFS volume, you are really striping similar vdevs. Here are some examples:

  • to extend a ZFS stripe, add one or more disks. Since there is no redundancy, you do not have to add the same amount of disks as the existing stripe.
  • to extend a ZFS mirror, add the same number of drives. The resulting striped mirror is a RAID 10. For example, if you have 10 drives, you could start by creating a mirror of two drives, extending this mirror by creating another mirror of two drives, and repeating three more times until all 10 drives have been added.
  • to extend a three drive RAIDZ1, add three additional drives. The result is a RAIDZ+0, similar to RAID 50 on a hardware controller.
  • to extend a RAIDZ2 requires a minimum of four additional drives. The result is a RAIDZ2+0, similar to RAID 60 on a hardware controller.

If you try to add an incorrect number of disks to the existing vdev, an error message will appear, indicating the number of disks that are needed. You will need to select the correct number of disks in order to continue.

8.1.1.3.1. Adding L2ARC or ZIL Devices

Storage Volumes Volume Manager (see Figure 8.1.1) is also used to add L2ARC or ZIL SSDs to improve specific types of volume performance. This is described in more detail in the ZFS Primer.

After the SSDs have been physically installed, click the Volume Manager button and choose the volume from the Volume to extend drop-down menu. Click the + next to the SSD in the Available disks list. In the Volume layout drop-down menu, select Cache (L2ARC) to add a cache device, or Log (ZIL) to add a log device. Finally, click Extend Volume to add the SSD.

8.1.2. Change Permissions

Setting permissions is an important aspect of configuring volumes. The graphical administrative interface is meant to set the initial permissions for a volume or dataset in order to make it available as a share. Once a share is available, the client operating system should be used to fine-tune the permissions of the files and directories that are created by the client.

The chapter on Sharing contains configuration examples for several types of permission scenarios. This section provides an overview of the screen that is used to set permissions.

Note

For users and groups to be available, they must either be first created using the instructions in Account or imported from a directory service using the instructions in Directory Services. If more than 50 users or groups are available, the drop-down menus described in this section will automatically truncate their display to 50 for performance reasons. In this case, start to type in the desired user or group name so that the display narrows its search to matching results.

After a volume or dataset is created, it is listed by its mount point name in Storage Volumes. Clicking the Change Permissions icon for a specific volume/dataset displays the screen shown in Figure 8.1.3. Table 8.1.3 summarizes the options in this screen.

_images/perms1.png

Fig. 8.1.3 Changing Permissions on a Volume or Dataset

Table 8.1.3 Options When Changing Permissions
Setting Value Description
Apply Owner (user) checkbox uncheck to prevent new permission change from being applied to Owner (user), see Note below
Owner (user) drop-down menu user to control the volume/dataset; users which were manually created or imported from a directory service will appear in the drop-down menu
Apply Owner (group) checkbox uncheck to prevent new permission change from being applied to Owner (group), see Note below
Owner (group) drop-down menu group to control the volume/dataset; groups which were manually created or imported from a directory service will appear in the drop-down menu
Apply Mode checkbox uncheck to prevent new permission change from being applied to Mode, see Note below
Mode checkboxes only applies to the Unix or Mac “Permission Type” so will be grayed out if Windows is selected
Permission Type bullet selection choices are Unix, Mac or Windows; select the type which matches the type of client accessing the volume/dataset
Set permission recursively checkbox if checked, permissions will also apply to subdirectories of the volume/dataset; if data already exists on the volume/dataset, change the permissions on the client side to prevent a performance lag

Note

The Apply Owner (user), Apply Owner (group), and Apply Mode checkboxes allow fine-tuning of the change permissions behavior. By default, all boxes are checked and FreeNAS® resets the owner, group, and mode when the Change button is clicked. These checkboxes allow choosing which settings to change. For example, to change just the Owner (group) setting, uncheck the boxes Apply Owner (user) and Apply Mode.

The Windows Permission Type is used for SMB shares or when the FreeNAS® system is a member of an Active Directory domain. This adds ACLs to traditional Unix permissions. When the Windows Permission Type is set, ACLs are set to Windows defaults for new files and directories. A Windows client can be used to further fine-tune permissions as needed.

The Unix Permission Type is usually used with NFS shares. These permissions are compatible with most network clients and generally work well with a mix of operating systems or clients. However, Unix permissions do not support Windows ACLs and should not be used with SMB shares.

The Mac Permission Type is used with AFP shares.

After a volume or dataset has been set to Windows, it cannot be changed to Unix permissions because that would remove extended permissions provided by Windows ACLs.

8.1.3. Create Dataset

An existing ZFS volume can be divided into datasets. Permissions, compression, deduplication, and quotas can be set on a per-dataset basis, allowing more granular control over access to storage data. A dataset is similar to a folder in that you can set permissions; it is also similar to a filesystem in that you can set properties such as quotas and compression as well as create snapshots.

Note

ZFS provides thick provisioning using quotas and thin provisioning using reserved space.

Selecting an existing ZFS volume in the tree and clicking Create Dataset shows the screen in Figure 8.1.4.

_images/dataset1b.png

Fig. 8.1.4 Creating a ZFS Dataset

Table 8.1.4 summarizes the options available when creating a ZFS dataset. Some settings are only available in Advanced Mode. To see these settings, either click the Advanced Mode button, or configure the system to always display these settings by checking the box Show advanced fields by default in System Advanced. Most attributes, except for the Dataset Name, Case Sensitivity, and Record Size, can be changed after dataset creation by highlighting the dataset name and clicking its Edit Options button in Storage Volumes.

Table 8.1.4 ZFS Dataset Options
Setting Value Description
Dataset Name string mandatory; enter a unique name for the dataset
Comments string short comments or user notes about this dataset
Compression Level drop-down menu see the section on Compression for a description of the available algorithms
Share type drop-down menu select the type of share that will be used on the dataset; choices are UNIX for an NFS share, Windows for a SMB share, or Mac for an AFP share
Case Sensitivity drop-down menu choices are sensitive (default, assumes filenames are case sensitive), insensitive (assumes filenames are not case sensitive), or mixed (understands both types of filenames)
Enable atime Inherit, On, or Off controls whether the access time for files is updated when they are read; setting this property to Off avoids producing log traffic when reading files and can result in significant performance gains
Quota for this dataset integer only available in Advanced Mode; default of 0 disables quotas; specifying a value means to use no more than the specified size and is suitable for user datasets to prevent users from hogging available space
Quota for this dataset and all children integer only available in Advanced Mode; a specified value applies to both this dataset and any child datasets
Reserved space for this dataset integer only available in Advanced Mode; default of 0 is unlimited; specifying a value means to keep at least this much space free and is suitable for datasets containing logs which could take up all available free space
Reserved space for this dataset and all children integer only available in Advanced Mode; a specified value applies to both this dataset and any child datasets
ZFS Deduplication drop-down menu read the section on Deduplication before making a change to this setting
Record Size drop-down menu only available in Advanced Mode; while ZFS automatically adapts the record size dynamically to adapt to data, if the data has a fixed size (e.g. a database), matching that size may result in better performance

After a dataset is created, you can click on that dataset and select Create Dataset, thus creating a nested dataset, or a dataset within a dataset. A zvol can also be created within a dataset. When creating datasets, double-check that you are using the Create Dataset option for the intended volume or dataset. If you get confused when creating a dataset on a volume, click all existing datasets to close them–the remaining Create Dataset will be for the volume.

8.1.3.1. Deduplication

Deduplication is the process of ZFS transparently reusing a single copy of duplicated data to save space. Depending on the amount of duplicate data, deduplicaton can improve storage capacity, as less data is written and stored. However, deduplication is RAM intensive. A general rule of thumb is 5 GB of RAM per terabyte of deduplicated storage. In most cases, compression provides storage gains comparable to deduplication with less impact on performance.

In FreeNAS®, deduplication can be enabled during dataset creation. Be forewarned that there is no way to undedup the data within a dataset once deduplication is enabled, as disabling deduplication has NO EFFECT on existing data. The more data written to a deduplicated dataset, the more RAM it requires. When the system starts storing the DDTs (dedup tables) on disk because they no longer fit into RAM, performance craters. Further, importing an unclean pool can require between 3-5 GB of RAM per terabyte of deduped data, and if the system does not have the needed RAM, it will panic. The only solution is to add more RAM or recreate the pool. Think carefully before enabling dedup! This article provides a good description of the value versus cost considerations for deduplication.

Unless a lot of RAM and a lot of duplicate data is available, do not change the default deduplication setting of “Off”. For performance reasons, consider using compression rather than turning this option on.

If deduplication is changed to On, duplicate data blocks are removed synchronously. The result is that only unique data is stored and common components are shared among files. If deduplication is changed to Verify, ZFS will do a byte-to-byte comparison when two blocks have the same signature to make sure that the block contents are identical. Since hash collisions are extremely rare, Verify is usually not worth the performance hit.

Note

After deduplication is enabled, the only way to disable it is to use the zfs set dedup=off dataset_name command from Shell. However, any data that has already been deduplicated will not be un-deduplicated. Only newly stored data after the property change will not be deduplicated. The only way to remove existing deduplicated data is to copy all of the data off of the dataset, set the property to off, then copy the data back in again. Alternately, create a new dataset with ZFS Deduplication left disabled, copy the data to the new dataset, and destroy the original dataset.

Tip

Deduplication is often considered when using a group of very similar virtual machine images. However, other features of ZFS can provide dedup-like functionality more efficiently. For example, create a dataset for a standard VM, then clone that dataset for other VMs. Only the difference between each created VM and the main dataset are saved, giving the effect of deduplication without the overhead.

8.1.3.2. Compression

When selecting a compression type, you need to balance performance with the amount of disk space saved by compression. Compression is transparent to the client and applications as ZFS automatically compresses data as it is written to a compressed dataset or zvol and automatically decompresses that data as it is read. These compression algorithms are supported:

  • lz4: recommended compression method as it allows compressed datasets to operate at near real-time speed. This algorithm only compresses the files that will benefit from compression. By default, ZFS pools made using FreeNAS® 9.2.1 or higher use this compression method, meaning that this algorithm is used if the Compression level is left at Inherit when creating a dataset or zvol.
  • gzip: varies from levels 1 to 9 where gzip fastest (level 1) gives the least compression and gzip maximum (level 9) provides the best compression but is discouraged due to its performance impact.
  • zle: fast but simple algorithm to eliminate runs of zeroes.
  • lzjb: provides decent data compression, but is considered deprecated as lz4 provides much better performance.

If you select Off as the Compression level when creating a dataset or zvol, compression will not be used on the dataset/zvol. This is not recommended as using lz4 has a negligible performance impact and allows for more storage capacity.

8.1.4. Create zvol

A zvol is a feature of ZFS that creates a raw block device over ZFS. This allows you to use a zvol as an iSCSI device extent.

To create a zvol, select an existing ZFS volume or dataset from the tree then click Create zvol to open the screen shown in Figure 8.1.5.

_images/zvol1a.png

Fig. 8.1.5 Creating a zvol

The configuration options are described in Table 8.1.5. Some settings are only available in Advanced Mode. To see these settings, either click the Advanced Mode button or configure the system to always display these settings by checking Show advanced fields by default in System Advanced.

Table 8.1.5 zvol Configuration Options
Setting Value Description
zvol Name string mandatory; enter a name for the zvol; note that there is a 63-character limit on device path names in devfs, so using long zvol names can prevent accessing zvols as devices; for example, a zvol with a 70-character filename or path cannot be used as an iSCSI extent
Comments string short comments or user notes about this zvol
Size for this zvol integer specify size and value such as 10Gib; if the size is more than 80% of the available capacity, the creation will fail with an “out of space” error unless Force size is checked
Force size checkbox by default, the system will not let you create a zvol if that operation will bring the pool to over 80% capacity; while NOT recommended, checking this box will force the creation of the zvol in this situation
Compression level drop-down menu see the section on Compression for a description of the available algorithms
Sparse volume checkbox used to provide thin provisioning; use with caution for when this option is selected, writes will fail when the pool is low on space
Block size drop-down menu only available in Advanced Mode and by default is based on the number of disks in pool; can be set to match the block size of the filesystem which will be formatted onto the iSCSI target

8.1.5. Import Disk

The Volume Import Disk screen, shown in Figure 8.1.6, is used to import a single disk that has been formatted with the UFS, NTFS, MSDOS, or EXT2 filesystem. The import is meant to be a temporary measure to copy the data from a disk to an existing ZFS dataset. Only one disk can be imported at a time.

Note

Imports of EXT3 or EXT4 filesystems are possible in some cases, although neither is fully supported. EXT3 journaling is not supported, so those filesystems must have an external fsck utility, like the one provided by E2fsprogs utilities, run on them before import. EXT4 filesystems with extended attributes or inodes greater than 128 bytes are not supported. EXT4 filesystems with EXT3 journaling must have an fsck run on them before import, as described above.

_images/import1.png

Fig. 8.1.6 Importing a Disk

Use the drop-down menu to select the disk to import, select the type of filesystem on the disk, and browse to the ZFS dataset that will hold the copied data. When you click Import Volume, the disk is mounted, its contents are copied to the specified ZFS dataset, and the disk is unmounted after the copy operation completes.

8.1.6. Import Volume

If you click Storage Volumes Import Volume, you can configure FreeNAS® to use an existing ZFS pool. This action is typically performed when an existing FreeNAS® system is re-installed. Since the operating system is separate from the storage disks, a new installation does not affect the data on the disks. However, the new operating system needs to be configured to use the existing volume.

Figure 8.1.7 shows the initial pop-up window that appears when you import a volume.

_images/auto1.png

Fig. 8.1.7 Initial Import Volume Screen

If you are importing an unencrypted ZFS pool, select No: Skip to import to open the screen shown in Figure 8.1.8.

_images/auto2.png

Fig. 8.1.8 Importing a Non-Encrypted Volume

Existing volumes should be available for selection from the drop-down menu. In the example shown in Figure 8.1.8, the FreeNAS® system has an existing, unencrypted ZFS pool. Once the volume is selected, click the OK button to import the volume.

If an existing ZFS pool does not show in the drop-down menu, run zpool import from Shell to import the pool.

If you plan to physically install ZFS formatted disks from another system, be sure to export the drives on that system to prevent an “in use by another machine” error during the import.

If you suspect that your hardware is not being detected, run camcontrol devlist from Shell. If the disk does not appear in the output, check to see if the controller driver is supported or if it needs to be loaded using Tunables.

8.1.6.1. Importing an Encrypted Pool

If you are importing an existing GELI-encrypted ZFS pool, you must decrypt the disks before importing the pool. In Figure 8.1.7, select Yes: Decrypt disks to access the screen shown in Figure 8.1.9.

_images/decrypt.png

Fig. 8.1.9 Decrypting Disks Before Importing a ZFS Pool

Select the disks in the encrypted pool, browse to the location of the saved encryption key, input the passphrase associated with the key, then click OK to decrypt the disks.

Note

The encryption key is required to decrypt the pool. If the pool cannot be decrypted, it cannot be re-imported after a failed upgrade or lost configuration. This means that it is very important to save a copy of the key and to remember the passphrase that was configured for the key. Refer to Managing Encrypted Volumes for instructions on how to manage the keys for encrypted volumes.

Once the pool is decrypted, it will appear in the drop-down menu of Figure 8.1.8. Click the OK button to finish the volume import.

8.1.7. View Disks

Storage Volumes View Disks shows all of the disks recognized by the FreeNAS® system. An example is shown in Figure 8.1.10.

_images/view.png

Fig. 8.1.10 Viewing Disks

The current configuration of each device is displayed. Click a disk entry and the Edit button to change its configuration. The configurable options are described in Table 8.1.6.

Table 8.1.6 Disk Options
Setting Value Description
Name string read-only value showing FreeBSD device name for disk
Serial string read-only value showing the disk’s serial number
Description string optional
HDD Standby drop-down menu indicates the time of inactivity (in minutes) before the drive enters standby mode in order to conserve energy; this forum post demonstrates how to determine if a drive has spun down
Advanced Power Management drop-down menu default is Disabled, can select a power management profile from the menu
Acoustic Level drop-down menu default is Disabled; can be modified for disks that understand AAM
Enable S.M.A.R.T. checkbox enabled by default if the disk supports S.M.A.R.T.; unchecking this box will disable any configured S.M.A.R.T. Tests for the disk
S.M.A.R.T. extra options string additional smartctl(8) options

Note

If a disk’s serial number is not displayed in this screen, use the smartctl command from Shell. For example, to determine the serial number of disk ada0, type smartctl -a /dev/ada0 | grep Serial.

The Wipe function is provided for when an unused disk is to be discarded.

Warning

Make certain that all data has been backed up and that the disk is no longer in use. Triple-check that the correct disk is being selected to be wiped, as recovering data from a wiped disk is usually impossible. If there is any doubt, physically remove the disk, verify that all data is still present on the FreeNAS® system, and wipe the disk in a separate computer.

Clicking Wipe offers several choices. Quick erases only the partitioning information on a disk, making it easy to reuse but without clearing other old data. For more security, Full with zeros overwrites the entire disk with zeros, while Full with random data overwrites the entire disk with random binary data.

Quick wipes take only a few seconds. A Full with zeros wipe of a large disk can take several hours, and a Full with random data takes longer. A progress bar is displayed during the wipe to track status.

8.1.8. Volumes

Storage Volumes is used to view and further configure existing ZFS pools, datasets, and zvols. The example shown in Figure 8.1.11 shows one ZFS pool (volume1) with two datasets (the one automatically created with the pool, volume1, and dataset1) and one zvol (zvol1).

Note that in this example, there are two datasets named volume1. The first represents the ZFS pool and its Used and Available entries reflect the total size of the pool, including disk parity. The second represents the implicit or root dataset and its Used and Available entries indicate the amount of disk space available for storage.

Buttons are provided for quick access to Volume Manager, Import Disk, Import Volume, and View Disks. If the system has multipath-capable hardware, an extra button will be added, View Multipaths. For each entry, the columns indicate the Name, how much disk space is Used, how much disk space is Available, the type of Compression, the Compression Ratio, the Status, whether it is mounted as read-only, and any Comments entered for the volume.

_images/volume1c.png

Fig. 8.1.11 Viewing Volumes

Clicking the entry for a pool causes several buttons to appear at the bottom of the screen. The buttons perform these actions:

Detach Volume: allows you to either export the pool or to delete the contents of the pool, depending upon the choice you make in the screen shown in Figure 8.1.12. The Detach Volume screen displays the current used space and indicates if there are any shares, provides checkboxes to Mark the disks as new (destroy data) and to Also delete the share’s configuration, asks if you are sure that you want to do this, and the browser will turn red to alert you that you are about to do something that will make the data inaccessible. If you do not check the box to mark the disks as new, the volume will be exported. This means that the data is not destroyed and the volume can be re-imported at a later time. If you will be moving a ZFS pool from one system to another, perform this export action first as it flushes any unwritten data to disk, writes data to the disk indicating that the export was done, and removes all knowledge of the pool from the system. If you do check the box to mark the disks as new, the pool and all the data in its datasets, zvols, and shares will be destroyed and the underlying disks will be returned to their raw state.

_images/detach1.png

Fig. 8.1.12 Detach or Delete a Volume

Scrub Volume: scrubs and scheduling them are described in more detail in Scrubs. This button allows manually initiating a scrub. Scrubs are I/O intensive and can negatively impact performance. Avoid initiating a scrub when the system is busy.

A Cancel button is provided to cancel a scrub. When a scrub is cancelled, it is abandoned. The next scrub to run starts from the beginning, not where the cancelled scrub left off.

The status of a running scrub or the statistics from the last completed scrub can be seen by clicking the Volume Status button.

Volume Status: as shown in the example in Figure 8.1.13, this screen shows the device name and status of each disk in the ZFS pool as well as any read, write, or checksum errors. It also indicates the status of the latest ZFS scrub. Clicking the entry for a device causes buttons to appear to edit the device’s options (shown in Figure 8.1.14), offline or online the device, or replace the device (as described in Replacing a Failed Drive).

Upgrade: used to upgrade the pool to the latest ZFS features, as described in Upgrading a ZFS Pool. This button does not appear if the pool is running the latest version of feature flags.

_images/volume2.png

Fig. 8.1.13 Volume Status

Selecting a disk in Volume Status and clicking its Edit Disk button shows the screen in Figure 8.1.14. Table 8.1.6 summarizes the configurable options.

_images/disk.png

Fig. 8.1.14 Editing a Disk

Note

Versions of FreeNAS® prior to 8.3.1 required a reboot to apply changes to the HDD Standby, Advanced Power Management, and Acoustic Level settings. As of 8.3.1, changes to these settings are applied immediately.

Clicking a dataset in Storage Volumes causes buttons to appear at the bottom of the screen, providing these options:

Change Permissions: edit the dataset’s permissions as described in Change Permissions.

Create Snapshot: create a one-time snapshot. To schedule the regular creation of snapshots, instead use Periodic Snapshot Tasks.

Destroy Dataset: clicking the Destroy Dataset button causes the browser window to turn red to indicate that this is a destructive action. The Destroy Dataset screen forces you to check the box I’m aware this will destroy all child datasets and snapshots within this dataset before it will perform this action.

Edit Options: edit the volume’s properties described in Table 8.1.4. Note that it will not allow changing the dataset’s name.

Create Dataset: used to create a child dataset within this dataset.

Create zvol: create a child zvol within this dataset.

Clicking a zvol in Storage Volumes causes icons to appear at the bottom of the screen: Create Snapshot, Edit zvol, and Destroy zvol. Similar to datasets, a zvol’s name cannot be changed, and destroying a zvol requires confirmation.

8.1.8.1. Managing Encrypted Volumes

If the Encryption box is checked during the creation of a pool, additional buttons appear in the entry for the volume in Storage Volumes. An example is shown in Figure 8.1.15.

_images/encrypt1.png

Fig. 8.1.15 Encryption Icons Associated with an Encrypted Volume

These additional encryption buttons are used to:

Create/Change Passphrase: set and confirm a passphrase associated with the GELI encryption key. The desired passphrase is entered and repeated for verification. A red warning is a reminder to Remember to add a new recovery key as this action invalidates the previous recovery key. Unlike a password, a passphrase can contain spaces and is typically a series of words. A good passphrase is easy to remember (like the line to a song or piece of literature) but hard to guess (people who know you should not be able to guess the passphrase). Remember this passphrase. An encrypted volume cannot be reimported without it. In other words, if the passphrase is forgotten, the data on the volume can become inaccessible if it becomes necessary to reimport the pool. Protect this passphrase, as anyone who knows it could reimport the encrypted volume, thwarting the reason for encrypting the disks in the first place.

_images/encrypt-passphrase.png

Fig. 8.1.16 Add or Change a Passphrase to an Encrypted Volume

After the passphrase is set, the name of this button changes to Change Passphrase. After setting or changing the passphrase, it is important to immediately create a new recovery key by clicking the Add recovery key button. This way, if the passphrase is forgotten, the associated recovery key can be used instead.

Encrypted volumes with a passphrase display an additional lock button:

_images/encrypt-lock.png

Fig. 8.1.17 Lock Button

These encrypted volumes can be locked. The data is not accessible until the volume is unlocked by suppying the passphrase or encryption key, and the button changes to an unlock button:

_images/encrypt-unlock.png

Fig. 8.1.18 Unlock Button

To unlock the volume, click the unlock button to display the Unlock dialog:

_images/encrypt-unlock-dialog.png

Fig. 8.1.19 Unlock Locked Volume

Unlock the volume by entering a passphrase or using the Browse button to load the recovery key. If both a passphrase and a recovery key are entered, only the passphrase is used. By default, the services listed will restart when the volume is unlocked. This allows them to see the new volume and share or access data on it. Individual services can be prevented from restarting by unchecking them. However, a service that is not restarted might not be able to access the unlocked volume.

Download Key: download a backup copy of the GELI encryption key. The encryption key is saved to the client system, not on the FreeNAS® system. The FreeNAS® administrative password must be entered, then the directory in which to store the key is chosen. Since the GELI encryption key is separate from the FreeNAS® configuration database, it is highly recommended to make a backup of the key. If the key is ever lost or destroyed and there is no backup key, the data on the disks is inaccessible.

Encryption Re-key: generate a new GELI encryption key. Typically this is only performed when the administrator suspects that the current key may be compromised. This action also removes the current passphrase.

Add recovery key: generate a new recovery key. This screen prompts for the FreeNAS® administrative password and then the directory in which to save the key. Note that the recovery key is saved to the client system, not on the FreeNAS® system. This recovery key can be used if the passphrase is forgotten. Always immediately add a recovery key whenever the passphrase is changed.

Remove recovery key: Typically this is only performed when the administrator suspects that the current recovery key may be compromised. Immediately create a new passphrase and recovery key.

Note

The passphrase, recovery key, and encryption key must be protected. Do not reveal the passphrase to others. On the system containing the downloaded keys, take care that the system and its backups are protected. Anyone who has the keys has the ability to re-import the disks if they are discarded or stolen.

Warning

If a re-key fails on a multi-disk system, an alert is generated. Do not ignore this alert as doing so may result in the loss of data.

8.1.9. View Multipaths

FreeNAS® uses gmultipath(8) to provide multipath I/O support on systems containing hardware that is capable of multipath. An example would be a dual SAS expander backplane in the chassis or an external JBOD.

Multipath hardware adds fault tolerance to a NAS as the data is still available even if one disk I/O path has a failure.

FreeNAS® automatically detects active/active and active/passive multipath-capable hardware. Any multipath-capable devices that are detected will be placed in multipath units with the parent devices hidden. The configuration will be displayed in Storage Volumes View Multipaths. Note that this option is not be displayed in the Storage Volumes tree on systems that do not contain multipath-capable hardware.

8.1.10. Replacing a Failed Drive

With any form of redundant RAID, failed drives must be replaced as soon as possible to repair the degraded state of the RAID. Depending on the hardware’s capabilities, it might be necessary to reboot to replace the failed drive. Hardware that supports AHCI does not require a reboot.

Note

Striping (RAID0) does not provide redundancy. If a disk in a stripe fails, the volume will be destroyed and must be recreated and the data restored from backup.

Note

If the volume is encrypted with GELI, refer to Replacing an Encrypted Drive before proceeding.

Before physically removing the failed device, go to Storage Volumes. Select the volume’s name. At the bottom of the interface are several icons, one of which is Volume Status. Click the Volume Status icon and locate the failed disk. Then perform these steps:

  1. Click the disk’s entry, then its Offline button to change the disk status to OFFLINE. This step is needed to properly remove the device from the ZFS pool and to prevent swap issues. If the hardware supports hot-pluggable disks, click the disk’s Offline button and pull the disk, then skip to step 3. If there is no Offline button but only a Replace button, the disk is already offlined and this step can be skipped.

    Note

    If the process of changing the disk’s status to OFFLINE fails with a “disk offline failed - no valid replicas” message, the ZFS volume must be scrubbed first with the Scrub Volume button in Storage Volumes. After the scrub completes, try to Offline the disk again before proceeding.

  2. If the hardware is not AHCI capable, shut down the system to physically replace the disk. When finished, return to the GUI and locate the OFFLINE disk.

  3. After the disk has been replaced and is showing as OFFLINE, click the disk again and then click its Replace button. Select the replacement disk from the drop-down menu and click the Replace Disk button. After clicking the Replace Disk button, the ZFS pool starts to resilver and the status of the resilver is displayed.

  4. After the drive replacement process is complete, re-add the replaced disk in the S.M.A.R.T. Tests screen.

In the example shown in Figure 8.1.20, a failed disk is being replaced by disk ada5 in the volume named volume1.

_images/replace.png

Fig. 8.1.20 Replacing a Failed Disk

After the resilver is complete, Volume Status shows a Completed resilver status and indicates any errors. Figure 8.1.21 indicates that the disk replacement was successful in this example.

Note

A disk that is failing but has not completely failed can be replaced in place, without first removing it. Whether this is a good idea depends on the overall condition of the failing disk. A disk with a few newly-bad blocks that is otherwise functional can be left in place during the replacement to provide data redundancy. A drive that is experiencing continuous errors can actually slow down the replacement. In extreme cases, a disk with serious problems might spend so much time retrying failures that it could prevent the replacement resilvering from completing before another drive fails.

_images/replace2.png

Fig. 8.1.21 Disk Replacement is Complete

8.1.10.1. Replacing an Encrypted Drive

If the ZFS pool is encrypted, additional steps are needed when replacing a failed drive.

First, make sure that a passphrase has been set using the instructions in Encryption before attempting to replace the failed drive. Then, follow the steps 1 and 2 as described above. During step 3, you will be prompted to input and confirm the passphrase for the pool. Enter this information then click the Replace Disk button. Wait until the resilvering is complete.

Next, restore the encryption keys to the pool. If the following additional steps are not performed before the next reboot, access to the pool might be permanently lost.

  1. Highlight the pool that contains the disk that was just replaced and click the Encryption Re-key button in the GUI. Entry of the root password will be required.
  2. Highlight the pool that contains the disk you just replaced and click Create Passphrase and enter the new passphrase. The old passphrase can be reused if desired.
  3. Highlight the pool that contains the disk you just replaced and click the Download Key button to save the new encryption key. Since the old key will no longer function, any old keys can be safely discarded.
  4. Highlight the pool that contains the disk that was just replaced and click the Add Recovery Key button to save the new recovery key. The old recovery key will no longer function, so it can be safely discarded.

8.1.10.2. Removing a Log or Cache Device

Added log or cache devices appear in Storage Volumes Volume Status. Clicking the device enables its Replace and Remove buttons.

Log and cache devices can be safely removed or replaced with these buttons. Both types of devices improve performance, and throughput can be impacted by their removal.

8.1.11. Replacing Drives to Grow a ZFS Pool

The recommended method for expanding the size of a ZFS pool is to pre-plan the number of disks in a vdev and to stripe additional vdevs using Volume Manager as additional capacity is needed.

However, this is not an option if there are no open drive ports and a SAS/SATA HBA card cannot be added. In this case, one disk at a time can be replaced with a larger disk, waiting for the resilvering process to incorporate the new disk into the pool, then repeating with another disk until all of the original disks have been replaced.

The safest way to perform this is to use a spare drive port or an eSATA port and a hard drive dock. The process follows these steps:

  1. Shut down the system.
  2. Install one new disk.
  3. Start up the system.
  4. Go to Storage Volumes, select the pool to expand and click the Volume Status button. Select a disk and click the Replace button. Choose the new disk as the replacement.
  5. The status of the resilver process can be viewed by running zpool status. When the new disk has resilvered, the old one will be automatically offlined. The system is then shut down to physically remove the replaced disk. One advantage of this approach is that there is no loss of redundancy during the resilver.

If a spare drive port is not available, a drive can be replaced with a larger one using the instructions in Replacing a Failed Drive. This process is slow and places the system in a degraded state. Since a failure at this point could be disastrous, do not attempt this method unless the system has a reliable backup. Replace one drive at a time and wait for the resilver process to complete on the replaced drive before replacing the next drive. After all the drives are replaced and the final resilver completes, the added space will appear in the pool.

8.2. Periodic Snapshot Tasks

A periodic snapshot task allows scheduling the creation of read-only versions of ZFS volumes and datasets at a given point in time. Snapshots can be created quickly and, if little data changes, new snapshots take up very little space. For example, a snapshot where no files have changed takes 0 MB of storage, but as changes are made to files, the snapshot size changes to reflect the size of the changes.

Snapshots provide a clever way of keeping a history of files, providing a way to recover an older copy or even a deleted file. For this reason, many administrators take snapshots often (perhaps every fifteen minutes), store them for a period of time (possibly a month), and store them on another system (typically using Replication Tasks). Such a strategy allows the administrator to roll the system back to a specific point in time. If there is a catastrophic loss, an off-site snapshot can be used to restore the system up to the time of the last snapshot.

An existing ZFS volume is required before creating a snapshot. Creating a volume is described in Volume Manager.

To create a periodic snapshot task, click Storage Periodic Snapshot Tasks Add Periodic Snapshot which opens the screen shown in Figure 8.2.1. Table 8.2.1 summarizes the fields in this screen.

Note

If only a one-time snapshot is needed, instead use Storage Volumes and click the Create Snapshot button for the volume or dataset to snapshot.

_images/periodic1b.png

Fig. 8.2.1 Creating a Periodic Snapshot

Table 8.2.1 Options When Creating a Periodic Snapshot
Setting Value Description
Volume/Dataset drop-down menu select an existing ZFS volume, dataset, or zvol
Recursive checkbox select this box to take separate snapshots of the volume/dataset and each of its child datasets; if unchecked, a single snapshot is taken of only the specified volume/dataset, but not any child datasets
Snapshot Lifetime integer and drop-down menu length of time to retain the snapshot on this system; if the snapshot is replicated, it is not removed from the receiving system when the lifetime expires
Begin drop-down menu do not create snapshots before this time of day
End drop-down menu do not create snapshots after this time of day
Interval drop-down menu how often to take snapshot between Begin and End times
Weekday checkboxes which days of the week to take snapshots
Enabled checkbox uncheck to disable the scheduled snapshot task without deleting it

If the Recursive box is checked, child datasets of this dataset are included in the snapshot and there is no need to create snapshots for each child dataset. The downside is that there is no way to exclude particular child datasets from a recursive snapshot.

When the OK button is clicked, a snapshot is taken and the task will be repeated according to your settings.

After creating a periodic snapshot task, an entry for the snapshot task will be added to View Periodic Snapshot Tasks. Click an entry to access its Edit and Delete buttons.

8.3. Replication Tasks

Replication is the duplication of snapshots from one FreeNAS® system to another computer. When a new snapshot is created on the source computer, it is automatically replicated to the destination computer. Replication is typically used to keep a copy of files on a separate system, with that system sometimes being at a different physical location.

The basic configuration requires a source system with the original data and a destination system where the data will be replicated. The destination system is prepared to receive replicated data, a periodic snapshot of the data on the source system is created, and then a replication task is created. As snapshots are automatically created on the source computer, they are automatically replicated to the destination computer.

Note

Replicated data is not visible on the receiving system until the replication task completes.

8.3.1. Examples: Common Configuration

The examples shown here use the same setup of source and destination computers.

8.3.1.1. Alpha (Source)

Alpha is the source computer with the data to be replicated. It is at IP address 10.0.0.102. A volume named alphavol has already been created, and a dataset named alphadata has been created on that volume. This dataset contains the files which will be snapshotted and replicated onto Beta.

This new dataset has been created for this example, but a new dataset is not required. Most users will already have datasets containing the data they wish to replicate.

Create a periodic snapshot of the source dataset by selecting Storage Periodic Snapshot Tasks. Click the alphavol/alphadata dataset to highlight it. Create a periodic snapshot of it by clicking Periodic Snapshot Tasks, then Add Periodic Snapshot as shown in Figure 8.3.1.

This example creates a snapshot of the alphavol/alphadata dataset every two hours from Monday through Friday between the hours of 9:00 and 18:00 (6:00 PM). Snapshots are automatically deleted after their chosen lifetime of two weeks expires.

_images/replication3.png

Fig. 8.3.1 Create a Periodic Snapshot for Replication

8.3.1.2. Beta (Destination)

Beta is the destination computer where the replicated data will be copied. It is at IP address 10.0.0.118. A volume named betavol has already been created.

Snapshots are transferred with SSH. To allow incoming connections, this service is enabled on Beta. The service is not required for outgoing connections, and so does not need to be enabled on Alpha.

8.3.2. Example: FreeNAS® to FreeNAS® Semi-Automatic Setup

FreeNAS® offers a special semi-automatic setup mode that simplifies setting up replication. Create the replication task on Alpha by clicking Replication Tasks and Add Replication. alphavol/alphadata is selected as the dataset to replicate. betavol is the destination volume where alphadata snapshots are replicated. The Setup mode dropdown is set to Semi-automatic as shown in Figure 8.3.2. The IP address of Beta is entered in the Remote hostname field. A hostname can be entered here if local DNS resolves for that hostname.

Note

If WebGUI HTTP –> HTTPS Redirect has been enabled in System General on the destination computer, Remote HTTP/HTTPS Port must be set to the HTTPS port (usually 443) and Remote HTTPS must be enabled when creating the replication on the source computer.

_images/replication6.png

Fig. 8.3.2 Add Replication Dialog, Semi-Automatic

The Remote Auth Token field expects a special token from the Beta computer. On Beta, choose Storage Replication Tasks, then click Temporary Auth Token. A dialog showing the temporary authorization token is shown as in Figure 8.3.3.

Highlight the temporary authorization token string with the mouse and copy it.

_images/replication7.png

Fig. 8.3.3 Temporary Authentication Token on Destination

On the Alpha system, paste the copied temporary authorization token string into the Remote Auth Token field as shown in Figure 8.3.4.

_images/replication8.png

Fig. 8.3.4 Temporary Authentication Token Pasted to Source

Finally, click the OK button to create the replication task. After each periodic snapshot is created, a replication task will copy it to the destination system. See Limiting Replication Times for information about restricting when replication is allowed to run.

Note

The temporary authorization token is only valid for a few minutes. If a Token is invalid message is shown, get a new temporary authorization token from the destination system, clear the Remote Auth Token field, and paste in the new one.

8.3.3. Example: FreeNAS® to FreeNAS® Dedicated User Replication

A dedicated user can be used for replications rather than the root user. This example shows the process using the semi-automatic replication setup between two FreeNAS® systems with a dedicated user named repluser. SSH key authentication is used to allow the user to log in remotely without a password.

In this example, the periodic snapshot task has not been created yet. If the periodic snapshot shown in the example configuration has already been created, go to Storage Periodic Snapshot Tasks, click on the task to select it, and click Delete to remove it before continuing.

On Alpha, select Account Users. Click the Add User. Enter repluser for Username, enter /mnt/alphavol/repluser in the Create Home Directory In field, enter Replication Dedicated User for the Full Name, and set the Disable password login checkbox. Leave the other fields at their default values, but note the User ID number. Click OK to create the user.

On Beta, the same dedicated user must be created as was created on the sending computer. Select Account Users. Click the Add User. Enter the User ID number from Alpha, repluser for Username, enter /mnt/betavol/repluser in the Create Home Directory In field, enter Replication Dedicated User for the Full Name, and set the Disable password login checkbox. Leave the other fields at their default values. Click OK to create the user.

A dataset with the same name as the original must be created on the destination computer, Beta. Select Storage Volumes, click on betavol, then click the Create Dataset icon at the bottom. Enter alphadata as the Dataset Name, then click Add Dataset.

The replication user must be given permissions to the destination dataset. Still on Beta, open a Shell and enter this command:

zfs allow -ldu repluser create,destroy,diff,mount,readonly,receive,release,send,userprop betavol/alphadata

The destination dataset must also be set to read-only. Enter this command in the Shell:

zfs set readonly=on betavol/alphadata

Close the Shell by typing exit and pressing Enter.

The replication user must also be able to mount datasets. Still on Beta, go to System Tunables. Click Add Tunable. Enter vfs.usermount for the Variable, 1 for the Value, and choose Sysctl from the Type drop-down. Click OK to save the tunable settings.

Back on Alpha, create a periodic snapshot of the source dataset by selecting Storage Periodic Snapshot Tasks. Click the alphavol/alphadata dataset to highlight it. Create a periodic snapshot of it by clicking Periodic Snapshot Tasks, then Add Periodic Snapshot as shown in Figure 8.3.1.

Still on Alpha, create the replication task by clicking Replication Tasks and Add Replication. alphavol/alphadata is selected as the dataset to replicate. betavol/alphadata is the destination volume and dataset where alphadata snapshots are replicated.

The Setup mode dropdown is set to Semi-automatic as shown in Figure 8.3.2. The IP address of Beta is entered in the Remote hostname field. A hostname can be entered here if local DNS resolves for that hostname.

Note

If WebGUI HTTP –> HTTPS Redirect has been enabled in System General on the destination computer, Remote HTTP/HTTPS Port must be set to the HTTPS port (usually 443) and Remote HTTPS must be enabled when creating the replication on the source computer.

The Remote Auth Token field expects a special token from the Beta computer. On Beta, choose Storage Replication Tasks, then click Temporary Auth Token. A dialog showing the temporary authorization token is shown as in Figure 8.3.3.

Highlight the temporary authorization token string with the mouse and copy it.

On the Alpha system, paste the copied temporary authorization token string into the Remote Auth Token field as shown in Figure 8.3.4.

Set the Dedicated User checkbox. Choose repluser in the Dedicated User drop-down.

Click the OK button to create the replication task.

Note

The temporary authorization token is only valid for a few minutes. If a Token is invalid message is shown, get a new temporary authorization token from the destination system, clear the Remote Auth Token field, and paste in the new one.

Replication will begin when the periodic snapshot task runs.

Additional replications can use the same dedicated user that has already been set up. The permissions and read only settings made through the Shell must be set on each new destination dataset.

8.3.4. Example: FreeNAS® to FreeNAS® or Other Systems, Manual Setup

This example uses the same basic configuration of source and destination computers shown above, but the destination computer is not required to be a FreeNAS® system. Other operating systems can receive the replication if they support SSH, ZFS, and the same features that are in use on the source system. The details of creating volumes and datasets, enabling SSH, and copying encryption keys will vary when the destination computer is not a FreeNAS® system.

8.3.4.1. Encryption Keys

A public encryption key must be copied from Alpha to Beta to allow a secure connection without a password prompt. On Alpha, select Storage Replication Tasks View Public Key, producing the window shown in Figure 8.3.5. Use the mouse to highlight the key data shown in the window, then copy it.

_images/replication1b.png

Fig. 8.3.5 Copy the Replication Key

On Beta, select Account Users View Users. Click the root account to select it, then click Modify User. Paste the copied key into the SSH Public Key field and click OK as shown in Figure 8.3.6.

_images/replication4.png

Fig. 8.3.6 Paste the Replication Key

Back on Alpha, create the replication task by clicking Replication Tasks and Add Replication. alphavol/alphadata is selected as the dataset to replicate. The destination volume is betavol. The alphadata dataset and snapshots are replicated there. The IP address of Beta is entered in the Remote hostname field as shown in Figure 8.3.7. A hostname can be entered here if local DNS resolves for that hostname.

Click the SSH Key Scan button to retrieve the SSH host keys from Beta and fill the Remote hostkey field. Finally, click OK to create the replication task. After each periodic snapshot is created, a replication task will copy it to the destination system. See Limiting Replication Times for information about restricting when replication is allowed to run.

_images/replication5.png

Fig. 8.3.7 Add Replication Dialog

8.3.5. Replication Options

Table 8.3.1 describes the options in the replication task dialog.

Table 8.3.1 Replication Task Options
Setting Value Description
Volume/Dataset drop-down menu ZFS volume or dataset on the source computer containing the snapshots to be replicated; the drop-down menu is empty if a snapshot does not already exist
Remote ZFS Volume/Dataset string ZFS volume on the remote or destination computer which will store the snapshots; if the destination dataset is not present, it will be created; /mnt/ is assumed, do not include it in the path
Recursively replicate child dataset’s snapshots checkbox when checked, also replicate snapshots of datasets that are children of the main dataset
Delete stale snapshots checkbox when checked, delete previous snapshots on the remote or destination computer which are no longer present on the source computer
Replication Stream Compression drop-down menu choices are lz4 (fastest), pigz (all rounder), plzip (best compression), or Off (no compression); selecting a compression algorithm can reduce the size of the data being replicated
Limit (kB/s) integer limit replication speed to the specified value in kilobytes/second; default of 0 is unlimited
Begin drop-down menu replication is not allowed to start before this time; times entered in the Begin and End fields set when replication can occur
End drop-down menu replication must start by this time; once started, replication will continue until it is finished
Enabled checkbox uncheck to disable the scheduled replication task without deleting it
Setup mode drop-down menu Manual or Semi-automatic
Remote hostname string IP address or DNS name of remote computer where replication is sent
Remote port string the port used by the SSH server on the remote or destination computer
Dedicated User Enabled checkbox allow a user account other than root to be used for replication
Dedicated User drop-down menu only available if Dedicated User Enabled is checked; select the user account to be used for replication
Encryption Cipher drop-down menu Standard, Fast, or Disabled
Remote hostkey string use the SSH Key Scan button to retrieve the public host key of the remote or destination computer and populate this field with that key

The replication task runs after a new periodic snapshot is created. The periodic snapshot and any new manual snapshots of the same dataset are replicated onto the destination computer.

When multiple replications have been created, replication tasks run serially, one after another. Completion time depends on the number and size of snapshots and the bandwidth available between the source and destination computers.

The first time a replication runs, it must duplicate data structures from the source to the destination computer. This can take much longer to complete than subsequent replications, which only send differences in data.

Selecting Storage Replication Tasks displays Figure 8.3.8, the list of replication tasks. The Last snapshot sent to remote side column shows the name of the last snapshot that was successfully replicated, and Status shows the current status of each replication task. The display is updated every five seconds, always showing the latest status.

_images/replication9.png

Fig. 8.3.8 Replication Task List

After a snapshot has been successfully replicated to another system, an OK is shown in the Replication column of the snapshot list at Storage Snapshots.

Note

The encryption key that was copied from the source computer (Alpha) to the destination computer (Beta) is an RSA public key located in the /data/ssh/replication.pub file on the source computer. The host public key used to identify the destination computer (Beta) is from the /etc/ssh/ssh_host_rsa_key.pub file on the destination computer.

8.3.6. Replication Encryption

The default Encryption Cipher Standard setting provides good security. Fast is less secure than Standard but can give reasonable transfer rates for devices with limited cryptographic speed. For networks where the entire path between source and destination computers is trusted, the Disabled option can be chosen to send replicated data without encryption.

8.3.7. Limiting Replication Times

The Begin and End times in a replication task make it possible to restrict when replication is allowed. These times can be set to only allow replication after business hours, or at other times when disk or network activity will not slow down other operations like snapshots or Scrubs. The default settings allow replication to occur at any time.

These times control when replication task are allowed to start, but will not stop a replication task that is already running. Once a replication task has begun, it will run until finished.

8.3.8. Troubleshooting Replication

Replication depends on SSH, disks, network, compression, and encryption to work. A failure or misconfiguration of any of these can prevent successful replication.

8.3.8.1. SSH

SSH must be able to connect from the source system to the destination system with an encryption key. This can be tested from Shell by making an SSH connection from the source system to the destination system. From the previous example, this is a connection from Alpha to Beta at 10.0.0.118. Start the Shell on the source machine (Alpha), then enter this command:

ssh -vv -i /data/ssh/replication 10.0.0.118

On the first connection, the system might say

No matching host key fingerprint found in DNS.
Are you sure you want to continue connecting (yes/no)?

Verify that this is the correct destination computer from the preceding information on the screen and type yes. At this point, an SSH shell connection is open to the destination system, Beta.

If a password is requested, SSH authentication is not working. See Figure 8.3.5 above. This key value must be present in the /root/.ssh/authorized_keys file on Beta, the destination computer. The /var/log/auth.log file can show diagnostic errors for login problems on the destination computer also.

8.3.8.2. Compression

Matching compression and decompression programs must be available on both the source and destination computers. This is not a problem when both computers are running FreeNAS®, but other operating systems might not have lz4, pigz, or plzip compression programs installed by default. An easy way to diagnose the problem is to set Replication Stream Compression to Off. If the replication runs, select the preferred compression method and check /var/log/debug.log on the FreeNAS® system for errors.

8.3.8.3. Manual Testing

On Alpha, the source computer, the /var/log/messages file can also show helpful messages to locate the problem.

On the source computer, Alpha, open a Shell and manually send a single snapshot to the destination computer, Beta. The snapshot used in this example is named auto-20161206.1110-2w. As before, it is located in the alphavol/alphadata dataset. A @ symbol separates the name of the dataset from the name of the snapshot in the command.

zfs send alphavol/alphadata@auto-20161206.1110-2w | ssh -i /data/ssh/replication 10.0.0.118 zfs recv betavol

If a snapshot of that name already exists on the destination computer, the system will refuse to overwrite it with the new snapshot. The existing snapshot on the destination computer can be deleted by opening a Shell on Beta and running this command:

zfs destroy -R betavol/alphadata@auto-20161206.1110-2w

Then send the snapshot manually again. Snapshots on the destination system, Beta, can be listed from the Shell with zfs list -t snapshot or by going to Storage Snapshots.

Error messages here can indicate any remaining problems.

8.4. Scrubs

A scrub is the process of ZFS scanning through the data on a volume. Scrubs help to identify data integrity problems, detect silent data corruptions caused by transient hardware issues, and provide early alerts of impending disk failures. FreeNAS® makes it easy to schedule periodic automatic scrubs.

Each volume should be scrubbed at least once a month. Bit errors in critical data can be detected by ZFS, but only when that data is read. Scheduled scrubs can find bit errors in rarely-read data. The amount of time needed for a scrub is proportional to the quantity of data on the volume. Typical scrubs take several hours or longer.

The scrub process is I/O intensive and can negatively impact performance. Schedule scrubs for evenings or weekends to minimize impact to users. Make certain that scrubs and other disk-intensive activity like S.M.A.R.T. Tests are scheduled to run on different days to avoid disk contention and extreme performance impacts.

Scrubs only check used disk space. To check unused disk space, schedule S.M.A.R.T. Tests of Type Long Self-Test to run once or twice a month.

Scrubs are scheduled and managed with Storage Scrubs.

When a volume is created, a ZFS scrub is automatically scheduled. An entry with the same volume name is added to Storage Scrubs. A summary of this entry can be viewed with Storage Scrubs View Scrubs. Figure 8.4.1 displays the default settings for the volume named volume1. In this example, the entry has been highlighted and the Edit button clicked to display the Edit screen. Table 8.4.1 summarizes the options in this screen.

_images/scrub1.png

Fig. 8.4.1 Viewing a Volume’s Default Scrub Settings

Table 8.4.1 ZFS Scrub Options
Setting Value Description
Volume drop-down menu select ZFS volume to scrub
Threshold days integer number of days since the last scrub completed before the next scrub can occur, regardless of the calendar schedule; the default is a multiple of 7 which should ensure that the scrub always occurs on the same day of the week
Description string optional
Minute slider or minute selections if use the slider, scrub occurs every N minutes; if use minute selections, scrub starts at the highlighted minutes
Hour slider or hour selections if use the slider, scrub occurs every N hours; if use hour selections, scrub occurs at the highlighted hours
Day of Month slider or month selections if use the slider, scrub occurs every N days; if use month selections, scrub occurs on the highlighted days of the selected months
Month checkboxes scrub occurs on the selected months
Day of week checkboxes scrub occurs on the selected days; default is Sunday to least impact users
Enabled checkbox uncheck to disable the scheduled scrub without deleting it

Review the default selections and, if necessary, modify them to meet the needs of the environment.

Scrubs can be deleted with the Delete button, but deleting a scrub is not recommended as a scrub provides an early indication of disk issues that could lead to a disk failure. If a scrub is too intensive for the hardware, consider unchecking the Enabled button for the scrub as a temporary measure until the hardware can be upgraded.

8.5. Snapshots

The Snapshots tab is used to review the listing of available snapshots. An example is shown in Figure 8.5.1.

Note

If snapshots do not appear, check that the current time configured in Periodic Snapshot Tasks does not conflict with the Begin, End, and Interval settings. If the snapshot was attempted but failed, an entry is added to /var/log/messages. This log file can be viewed in Shell.

_images/periodic3a.png

Fig. 8.5.1 Viewing Available Snapshots

The listing includes the name of the volume or dataset, the name of each snapshot, and the amount of used and referenced data.

Used is the amount of space consumed by this dataset and all of its descendants. This value is checked against the dataset’s quota and reservation. The space used does not include the dataset’s reservation, but does take into account the reservations of any descendent datasets. The amount of space that a dataset consumes from its parent, as well as the amount of space that are freed if this dataset is recursively destroyed, is the greater of its space used and its reservation. When a snapshot is created, the space is initially shared between the snapshot and the filesystem, and possibly with previous snapshots. As the filesystem changes, space that was previously shared becomes unique to the snapshot, and is counted in the snapshot’s space used. Additionally, deleting snapshots can increase the amount of space unique to (and used by) other snapshots. The amount of space used, available, or referenced does not take into account pending changes. While pending changes are generally accounted for within a few seconds, disk changes do not necessarily guarantee that the space usage information is updated immediately.

Tip

Space used by individual snapshots can be seen by running zfs list -t snapshot from Shell.

Refer indicates the amount of data accessible by this dataset, which may or may not be shared with other datasets in the pool. When a snapshot or clone is created, it initially references the same amount of space as the file system or snapshot it was created from, since its contents are identical.

Replication shows whether the snapshot has been replicated to a remote system.

Snapshots have icons on the right side for several actions.

Clone Snapshot prompts for the name of the clone to create. A clone is a writable copy of the snapshot. Since a clone is actually a dataset which can be mounted, it appears in the Volumes tab rather than the Snapshots tab. By default, -clone is added to the name of a snapshot when a clone is created.

Destroy Snapshot a pop-up message asks for confirmation. Child clones must be destroyed before their parent snapshot can be destroyed. While creating a snapshot is instantaneous, deleting a snapshot can be I/O intensive and can take a long time, especially when deduplication is enabled. In order to delete a block in a snapshot, ZFS has to walk all the allocated blocks to see if that block is used anywhere else; if it is not, it can be freed.

The most recent snapshot also has a Rollback Snapshot icon. Clicking the icon asks for confirmation before rolling back to this snapshot state. Confirming by clicking Yes causes any files that have changed since the snapshot was taken to be reverted back to their state at the time of the snapshot.

Note

Rollback is a potentially dangerous operation and causes any configured replication tasks to fail as the replication system uses the existing snapshot when doing an incremental backup. To restore the data within a snapshot, the recommended steps are:

  1. Clone the desired snapshot.
  2. Share the clone with the share type or service running on the FreeNAS® system.
  3. After users have recovered the needed data, destroy the clone in the Active Volumes tab.

This approach does not destroy any on-disk data and has no impact on replication.

A range of snapshots can be selected with the mouse. Click on the checkbox in the left column of the first snapshot, then press and hold Shift and click on the checkbox for the end snapshot. This can be used to select a range of obsolete snapshots to be deleted with the Destroy icon at the bottom. Be cautious and careful when deleting ranges of snapshots.

Periodic snapshots can be configured to appear as shadow copies in newer versions of Windows Explorer, as described in Configuring Shadow Copies. Users can access the files in the shadow copy using Explorer without requiring any interaction with the FreeNAS® graphical administrative interface.

The ZFS Snapshots screen allows the creation of filters to view snapshots by selected criteria. To create a filter, click the Define filter icon (near the text No filter applied). When creating a filter:

  • select the column or leave the default of Any Column.
  • select the condition. Possible conditions are: contains (default), is, starts with, ends with, does not contain, is not, does not start with, does not end with, and is empty.
  • enter a value that meets your view criteria.
  • click the Filter button to save your filter and exit the define filter screen. Alternately, click the + button to add another filter.

If you create multiple filters, select the filter to use before leaving the define filter screen. Once a filter is selected, the No filter applied text changes to Clear filter. If you click Clear filter, a pop-up message indicates that this removes the filter and all available snapshots are listed.

8.6. VMware-Snapshot

Storage VMware-Snapshot allows you to coordinate ZFS snapshots when using FreeNAS® as a VMware datastore. Once this type of snapshot is created, FreeNAS® will automatically snapshot any running VMware virtual machines before taking a scheduled or manual ZFS snapshot of the dataset or zvol backing that VMware datastore. The temporary VMware snapshots are then deleted on the VMware side but still exist in the ZFS snapshot and can be used as stable resurrection points in that snapshot. These coordinated snapshots will be listed in Snapshots.

Figure 8.6.1 shows the menu for adding a VMware snapshot and Table 8.6.1 summarizes the available options.

_images/vmware1a.png

Fig. 8.6.1 Adding a VMware Snapshot

Table 8.6.1 VMware Snapshot Options
Setting Value Description
Hostname string IP address or hostname of VMware host; when clustering, this is the vCenter server for the cluster
Username string user on VMware host with enough permission to snapshot virtual machines
Password string password associated with Username
ZFS Filesystem drop-down menu the filesystem to snapshot
Datastore drop-down menu after entering the Hostname, Username, and Password, click Fetch Datastores to populate the menu and select the datastore with which to synchronize