Dm-crypt

dm-crypt is Article description::a disk encryption system using the kernels crypto API framework and device mapper subsystem. With dm-crypt, administrators can encrypt entire disks, logical volumes, partitions, but also single files.

The dm-crypt subsystem supports the Linux Unified Key Setup (LUKS) structure, which allows for multiple keys to access the encrypted data, as well as manipulate the keys (such as changing the keys, adding additional passphrases, etc.) Although dm-crypt supports non-LUKS setups as well, this article will focus on the LUKS functionality mostly due to its flexibility, manageability as well as broad support in the community.

Configuration
There are two prerequisites before one can start using dm-crypt:


 * 1) Configuration of the Linux kernel
 * 2) Installation of the  package

Kernel Configuration
To use dm-crypt there are a number of configuration entries that are necessary.

First of all, support for the device mapper infrastructure as well as the crypt target must be included:

Next, the Linux kernel needs to support the set of cryptographic APIs that the administrator wants to use for encryption. These can be found under the Cryptographic API section:

If the root file system will be encrypted as well, then an initial ram file system needs to be created in which the root filesystem is decrypted before it is mounted. Thus this requires initramfs support as well:

If using the tcrypt encryption option (TrueCrypt/tcplay/VeraCrypt compatibility mode), then the following items will also need to be added to the kernel. Otherwise, cryptsetup will return the following errors: "device-mapper: reload ioctl failed: Invalid argument" and "Kernel doesn't support TCRYPT compatible mapping".

Cryptsetup installation
The package provides the  command, which is used to open or close the encrypted storage as well as manage the passphrases or keys associated with it.

Benchmark
provides a benchmarking tool which will help to decide which setup to choose. The output depends on kernel settings as well as USE flags and destination (HDD, SSD etc.).

Keyfile or passphrase
In order to start with encrypted storage, the administrator will need to decide which method to use for the encryption key. With the choice is either a passphrase or a keyfile. In case of a keyfile, this can be any file, but it is recommended to use a file with random data which is properly protected (considering that access to this keyfile will mean access to the encrypted data).

To create a keyfile, one can use the command:

In the next sections, we will show every command for both situations - passphrase and keyfile. Of course, only one method is necessary.

Creating an encrypted storage platform
In order to create an encrypted storage platform (which can be a disk, partition, file, ...) use the command with the   action.

For instance, to have as the storage medium for the encrypted data:

To use a keyfile instead of a passphrase:

The  tells  the cipher used to encrypt the disk (  will show you all possibilities). tells which keylength to use for the real encryption key (unlike the passphrase or keyfile, which are used to access this real encryption key). Finally  forces you to type your password twice.

Opening the encrypted storage
In order to open up the encrypted storage (i.e. make the real data accessible through transparent decryption), use the  action.

If a keyfile is used, then the command would look like so:

When the command finishes successfully, then a new device file called will be made available.

If this is the first time this encrypted device is used, it needs to be formatted. The following example uses the Btrfs file system but of course any other file system will do:

Once the file system is formatted, or the formatting was already done in the past, then the device file can be mounted on the system:

Closing the encrypted storage
In order to close the encrypted storage (i.e. ensure that the real data is no longer accessible through transparent decryption), use the  action:

Of course, make sure that the device is no longer in use.

Manipulating LUKS keys
LUKS keys are used to access the real encryption key. They are stored in slots in the header of the (encrypted) partition, disk or file.

Listing the slots
With the  action, information about the encrypted partition, disk or file can be shown. This includes the slots:

In the above example, two slots are used. Note that  does not give away anything sensitive - it is merely displaying the LUKS header content. No decryption key has to be provided in order to call.

Adding a keyfile or passphrase
In order to add an additional keyfile or passphrase to access the encrypted storage, use the  action:

To use a keyfile to unlock the key (but still add in a passphrase):

If a keyfile is to be added (say ):

Or, to use the first keyfile to unlock the main key:

Removing a keyfile or passphrase
With the  action, a keyfile or passphrase can be removed (so they can no longer be used to decrypt the storage):

Or to remove a keyfile:

Make sure that at least one method for accessing the data is still available. Once a passphrase or keyfile is removed for use, this cannot be recovered again.

Emptying a slot
Suppose the passphrase or keyfile is no longer known, then the slot can be freed. Of course, this does require prior knowledge of which slot that the passphrase or keyfile was stored in.

For instance, to empty out slot 2 (which is the third slot as slots are numbered starting from 0):

This command will ask for a valid passphrase before continuing. Or one can pass on the keyfile to use:

Automate mounting encrypted file systems
Until now, the article focused on manual setup and mounting/unmounting of encrypted file systems. An init service exists which automates the decrypting and mounting of encrypted file systems.

Configuring dm-crypt
Edit the file and add in entries for each file system. The supported entries are well documented in the file, the below example is just that - an example:

Configuring fstab
The next step is to configure to automatically mount the (decrypted) file systems when they become available. It is recommended to first obtain the UUID of the decrypted (mounted) file system:

Then, update the file accordingly:

Add initscript to bootlevel
Don't forget to have the init service launched at boot:

Make decrypted device nodes visible
If you have decrypted/unlocked a device before the services were started for example your root disk in an with an initramfs then it's possible that the mapped device is not visible. In this case you can run the following to recreate it.

Mounting TrueCrypt/tcplay/VeraCrypt volumes
Replace container-to-mount with the device file under or the path to the file you wish to open. Upon successful opening, the plaintext device will appear as, which you can  like any normal device.

If you are using key files, supply them using the  option, to open a hidden volume, supply the   option and for a partition or whole drive that is encrypted in system mode use the   option.

When done,  the volume, and close the container using the following command:

External resources

 * The cryptsetup FAQ hosted on GitLab covers a wide range of frequently asked questions.