TrueNAS SCALE Version DocumentationThis content follows the TrueNAS SCALE 24.04 (Dragonfish) releases. Use the Product and Version selectors above to view content specific to different TrueNAS software or major version.
Storage Encryption
13 minute read.
TrueNAS SCALE offers ZFS encryption for your sensitive data in pools and datasets or Zvols.
Users are responsible for backing up and securing encryption keys and passphrases! Losing the ability to decrypt data is similar to a catastrophic data loss.
Data-at-rest encryption is available with:
- Self Encrypting Drives (SEDs) using OPAL or FIPS 140.2 (Both AES 256)
- Encryption of specific datasets (AES-256-GCM)
The local TrueNAS system manages keys for data-at-rest. Users are responsible for storing and securing their keys. TrueNAS SCALE includes the Key Management Interface Protocol (KMIP).
Encryption is for users storing sensitive data. Pool-level encryption does not apply to the storage pool or the disks in the pool. It only applies to the root dataset that shares the same name as the pool. Child datasets or zvols inherit encryption from the parent dataset.
TrueNAS automatically generates a root dataset when you create a pool. This root dataset inherits the encryption state of the pool through the Encryption option on the Pool Creation Wizard screen when you create the pool. Because encryption is inherited from the parent, all data within that pool is encrypted. Selecting the Encryption option for the pool (root dataset) forces encryption for all datasets and zvols created within the root dataset.
You cannot create an unencrypted dataset within an encrypted pool or dataset. This change does not affect existing datasets created in earlier releases of SCALE but does affect new datasets created in 22.12.3 and later releases.
Leave the Encryption option on the Pool Creation Wizard screen cleared to create an unencrypted pool. You can create both unencrypted and encrypted datasets within an unencrypted pool (root dataset). If you create an encrypted dataset within an unencrypted dataset, all datasets or zvol created within that encrypted dataset are automatically encrypted.
If you have only one pool on your system, do not select the Encryption option for this pool.
If your system loses power or you reboot the system, the datasets, zvols, and all data in an encrypted pool automatically lock to protect the data in that encrypted pool.
SCALE uses lock icons to indicate the encryption state of a root, parent, or child dataset in the tree table on the Datasets screen. Each icon shows a text label with the state of the dataset when you hover the mouse over the icon.
The Datasets tree table includes lock icons and descriptions that indicate the encryption state of datasets.
Icon | State | Description |
---|---|---|
Locked | Displays for locked encrypted root, non-root parent and child datasets. | |
Unlocked | Displays for unlocked encrypted root, non-root parent and child datasets. | |
Locked by ancestor | Displays for locked datasets that inherit encryption properties from the parent. | |
Unlocked by ancestor | Displays for unlocked datasets that inherit encryption properties from the parent. |
A dataset that inherits encryption shows the mouse hover-over label Locked by ancestor or Unlocked by ancestor.
Select an encrypted dataset to see the ZFS Encryption widget on the Datasets screen.
The dataset encryption state is unlocked until you lock it using the Lock button on the ZFS Encryption widget. After locking the dataset, the icon on the tree table changes to locked, and the Unlock button appears on the ZFS Encryption widget.
Before creating a pool with encryption decide if you want to encrypt all datasets, zvols, and data stored on the pool.
If your system does not have enough disks to allow you to create a second storage pool, we recommend that you not use encryption at the pool level. Instead, apply encryption at the dataset level to non-root parent or child datasets. You can mix encrypted and unencrypted datasets on an unencrypted pool.You cannot change a pool from encrypted to non-encrypted. You can only change the dataset encryption type (key or passphrase) for the encrypted pool.
All pool-level encryption is key-based encryption. When prompted, download the encryption key and keep it stored in a safe place where you can back up the file. You cannot use passphrase encryption at the pool level.
Go to Storage and click Create Pool on the Storage Dashboard screen. You can also click Add to Pool on the Unassigned Disks widget and select the Add to New to open the Pool Creation Wizard.
Enter a name for the pool, select Encryption next to Name, then select the layout for the data VDEV and add the disks. A warning dialog displays after selecting Encryption.
Read the warning, select Confirm, and then click I UNDERSTAND.
A second dialog opens where you click Download Encryption Key for the pool encryption key.
Click Done to close the window. Move the encryption key to safe location where you can back up the file.
Add any other VDEVS to the pool you want to include, then click Save to create the pool with encryption.
To add an encrypted dataset, go to Datasets.
Select the dataset on the tree table where you want to add a new dataset. The default dataset selected when you open the Datasets screen is the root dataset of the first pool on the tree table list. If you have more than one pool and want to create a dataset in a pool other than the default, select the root dataset for that pool or any dataset under the root where you want to add the new dataset.
Click Add Dataset to open the Add Dataset screen, then click Advanced Options.
Enter a value in Name.
Select the Dataset Preset option you want to use. Options are:
- Generic for non-SMB share datasets such as iSCSI and NFS share datasets or datasets not associated with application storage.
- Multiprotocol for datasets optimized for SMB and NFS multi-mode shares or to create a dataset for NFS shares.
- SMB for datasets optimized for SMB shares.
- Apps for datasets optimized for application storage.
Generic sets ACL permissions equivalent to Unix permissions 755, granting the owner full control and the group and other users read and execute privileges.
SMB, Apps, and Multiprotocol inherit ACL permissions based on the parent dataset. If there is no ACL to inherit, one is calculated granting full control to the owner@, group@, members of the builtin_administrators group, and domain administrators. Modify control is granted to other members of the builtin_users group and directory services domain users.
Apps includes an additional entry granting modify control to group 568 (Apps).
To add encryption to a dataset, scroll down to Encryption Options and select the inherit checkbox to clear the checkmark. If the parent dataset is unencrypted and you want to encrypt the dataset, clear the checkmark to show the Encryption option. If the parent dataset is encrypted and you want to change the type, clearing the checkmark shows the other encryption options. To keep the dataset encryption settings from the parent, leave inherited checkmarked.
Decide if you want to use the default key type encryption and if you want to let the system generate the encryption key. To use key encryption and your own key, clear the Generate key checkbox to display the Key field. Enter your key in this field.
To change to passphrase encryption, click the down arrow and select Passphrase from the Encryption Type dropdown.
You can select the encryption algorithm to use from the Encryption Standard dropdown list of options or use the recommended default.
Leave the default selection if you do not have a particular encryption standard you want to use.
The passphrase must be longer than 8 and less than 512 characters.
Keep encryption keys and/or passphrases safeguarded in a secure and protected place. Losing encryption keys or passphrases can result in permanent data loss!
You cannot add encryption to an existing dataset. You can change the encryption type for an already encrypted dataset using the Edit option on the ZFS Encryption widget for the dataset.
Save any change to the encryption key or passphrase, and update your saved passcodes and keys file, and then back up that file.
To change the encryption type, go to Datasets:
Select the encrypted dataset on the tree table, then click Edit on the ZFS Encryption widget. The Edit Encryption Options dialog for the selected dataset displays.
You must unlock a locked encrypted dataset before you can make changes.
If the dataset inherits encryption settings from a parent dataset, to change this, clear the Inherit encryption properties from parent checkbox to display the key type encryption setting options.
If the encryption type is set to passphrase, you can change the passphrase, or change Encryption Type to key. You cannot change a dataset created with a key as the encryption type to passphrase.
Key type options are Generate Key (pre-selected) or clear to display the Key field. Enter your new key in this field.
To change the passphrase for passphrase-encryption, enter a new passphrase in Passphrase and Confirm Passphrase.
Use a complex passphrase that is not easy to guess. Store in a secure location subject to regular backups.
Leave the other settings at default, then click Confirm to activate Save.
Click Save to close the window and update the ZFS Encryption widget to reflect the changes made.
You can only lock and unlock an encrypted dataset if it is secured with a passphrase instead of a key file. Before locking a dataset, verify that it is not currently in use.
Select the encrypted dataset on the tree table, then click Lock on the ZFS Encryption widget to open the Lock Dataset dialog with the dataset full path name.
Use the Force unmount option only if you are certain no one is currently accessing the dataset. Force unmount boots anyone using the dataset (e.g. someone accessing a share) so you can lock it. Click Confirm to activate Lock, then click Lock.
You cannot use locked datasets.
To unlock a dataset, go to Datasets then select the locked dataset on the tree table. Click Unlock on the ZFS Encryption widget to open the Unlock Dataset screen.
Enter the key if key-encrypted, or the passphrase into Dataset Passphrase and click Save.
Select Unlock Child Encrypted Roots to unlock all locked child datasets if they use the same passphrase.
Select Force if the dataset mount path exists but is not empty. When this happens, the unlock operation fails. Using Force allows the system to rename the existing directory and file where the dataset should mount. This prevents the mount operation from failing. A confirmation dialog displays.
Click CONTINUE to confirm you want to unlock the datasets. Click CLOSE to exit and keep the datasets locked. A second confirmation dialog opens confirming the datasets unlocked. Click CLOSE. TrueNAS displays the dataset with the unlocked icon.
Encryption is for securing sensitive data.
You can only encrypt a Zvol if you create the Zvol from a dataset with encryption.
Users are responsible for backing up and securing encryption keys and passphrases! Losing the ability to decrypt data is similar to a catastrophic data loss.
Zvols inherit encryption settings from the parent dataset.
To encrypt a Zvol, select a dataset configured with encryption and then create a new Zvol.
Next, go to Datasets and click on the Zvol.
If you do not see the ZFS Encryption widget, you created the Zvol from an unencrypted dataset. Delete the Zvol and start over.
The Zvol is encrypted with settings inherited from the parent dataset.
To change inherited encryption properties from passphrase to key, or enter a new key or passphrase, select the zvol, then click Edit on the ZFS Encryption widget.
If Encryption Type is set to Key, type an encryption key into the Key field or select Generate Key. If using Passphrase, enter a passphrase of eight to 512 characters. Use a passphrase complex enough to not easily guess. After making any changes, select Confirm, and then click Save.
Save any change to the encryption key or passphrase, update your saved passcodes and keys file, and back up the file.
There are two ways to manage the encryption credentials, with a key file or passphrase. Creating a new encrypted pool automatically generates a new key file and prompts users to download it.
Always back up the key file to a safe and secure location.
To manually back up a root dataset key file, click Export Key on the ZFS Encryption widget.
See Changing Dataset-Level Encryption for more information on changing encryption settings.
A passphrase is a user-defined string of eight to 512 characters that is required to decrypt the dataset.
The pbkdf2iters is the number of password-based key derivation function 2 (PBKDF2) iterations to use for reducing vulnerability to brute-force attacks. Users must enter a number greater than 100000.
TrueNAS SCALE users should either replicate the dataset/Zvol without properties to disable encryption at the remote end or construct a special JSON manifest to unlock each child dataset/zvol with a unique key.
Replicate every encrypted dataset you want to replicate with properties.
Export key for every child dataset that has a unique key.
For each child dataset construct a proper json with poolname/datasetname of the destination system and key from the source system like this:
{"tank/share01": "57112db4be777d93fa7b76138a68b790d46d6858569bf9d13e32eb9fda72146b"}
Save this file with the extension
.json . On the remote system, unlock the dataset(s) using properly constructed
json files.
Uncheck properties when replicating so that the destination dataset is not encrypted on the remote side and does not require a key to unlock.
Go to Data Protection and click ADD in the Replication Tasks window.
Click Advanced Replication Creation.
Fill out the form as needed and make sure Include Dataset Properties is NOT checked.
Click Save.
Go to Datasets on the system you are replicating from. Select the dataset encrypted with a key, then click Export Key on the ZFS Encryption widget to export the key for the dataset.
Apply the JSON key file or key code to the dataset on the system you replicated the dataset to.
Option 1: Download the key file and open it in a text editor. Change the pool name/dataset part of the string to the pool name/dataset for the receiving system. For example, replicating from tank1/dataset1 on the replicate-from system to tank2/dataset2 on the replicate-to system.
Option 2: Copy the key code provided in the Key for dataset window.
On the system receiving the replicated pool/dataset, select the receiving dataset and click Unlock.
Unlock the dataset. Either clear the Unlock with Key file checkbox, paste the key code into the Dataset Key field (if there is a space character at the end of the key, delete the space), or select the downloaded Key file that you edited.
Click Save.
Click Continue.