| CRYPTTAB(5) | crypttab | CRYPTTAB(5) |
.SH "NAME" crypttab - Configuration for encrypted block devices
.SH "SYNOPSIS"
.PP /etc/crypttab
.SH "DESCRIPTION"
.PP The /etc/crypttab file describes encrypted block devices that are set up
during system boot.
.PP Empty lines and lines starting with the # character are ignored. Each of
the remaining lines describes one encrypted block device. Fields are
delimited by white space.
.PP Each line is in the form
volume-name encrypted-device key-file options
The first two fields are mandatory, the remaining two are optional.
.PP Setting up encrypted block devices using this file supports four
encryption modes: LUKS, TrueCrypt, BitLocker and plain. See
cryptsetup(8) for more information about each mode. When no mode is
specified in the options field and the block device contains a LUKS
signature, it is opened as a LUKS device; otherwise, it is assumed to be in
raw dm-crypt (plain mode) format.
.PP The four fields of /etc/crypttab are defined as follows:
If the specified key file path refers to an AF_UNIX stream socket in
the file system, the key is acquired by connecting to the socket and reading
it from the connection. This allows the implementation of a service to
provide key information dynamically, at the moment when it is needed. For
details see below.
.SH "KEY ACQUISITION"
.PP Six different mechanisms for acquiring the decryption key or passphrase
unlocking the encrypted volume are supported. Specifically:
.PP For the latter five mechanisms the source for the key material used for
unlocking the volume is primarily configured in the third field of each
/etc/crypttab line, but may also be configured in /etc/cryptsetup-keys.d/
and /run/cryptsetup-keys.d/ (see above) or in the LUKS2 JSON token header
(in case of the latter three). Use the systemd-cryptenroll(1) tool to
enroll PKCS#11, FIDO2 and TPM2 devices in LUKS2 volumes.
.SH "SUPPORTED OPTIONS"
.PP The following options may be used in the fourth field of each line:
.PP cipher=
Specifies the cipher to use. See cryptsetup(8) for possible values and
the default value of this option. A cipher with unpredictable IV values,
such as aes-cbc-essiv:sha256, is recommended. Embedded commas in the cipher
specification need to be escaped by preceding them with a backslash, see
example below.
Added in version 186.
.RE
.PP discard
Allow discard requests to be passed through the encrypted block device. This
improves performance on SSD storage but has security implications.
Added in version 207.
.RE
.PP hash=
Specifies the hash to use for password hashing. See cryptsetup(8) for
possible values and the default value of this option.
Added in version 186.
.RE
.PP header=
Use a detached (separated) metadata device or file where the header
containing the master key(s) is stored. This option is only relevant for
LUKS and TrueCrypt/VeraCrypt devices. See cryptsetup(8) for possible
values and the default value of this option.
Optionally, the path may be followed by : and an /etc/fstab device
specification (e.g. starting with UUID= or similar); in which case, the path
is relative to the device file system root. The device gets mounted
automatically for LUKS device activation duration only.
Added in version 219.
.RE
.PP keyfile-offset=
Specifies the number of bytes to skip at the start of the key file. See
cryptsetup(8) for possible values and the default value of this
option.
Added in version 187.
.RE
.PP keyfile-size=
Specifies the maximum number of bytes to read from the key file. See
cryptsetup(8) for possible values and the default value of this
option. This option is ignored in plain encryption mode, where the key file
size is determined by the key size. It is also ignored when the key file is
used as a salt file for a FIDO2 token, as the salt size in that case is
defined by the FIDO2 specification to be exactly 32 bytes.
Added in version 188.
.RE
.PP keyfile-erase
If enabled, the specified key file is erased after the volume is activated or
when activation fails. This is in particular useful when the key file is
only acquired transiently before activation (e.g. via a file in /run/,
generated by a service running before activation), and shall be removed
after use. Defaults to off.
Note that this option only applies to a key file explicitly configured in the
third field, and has no effect on key files that are automatically
discovered in /etc/cryptsetup-keys.d/ and /run/cryptsetup-keys.d/. The
latter are considered shared resources that are not owned by an individual
volume, and hence are never erased. To erase an automatically discovered key
file, configure its path explicitly in the third field.
Added in version 246.
.RE
.PP key-slot=
Specifies the key slot to compare the passphrase or key against. If the key
slot does not match the given passphrase or key, but another would, the
setup of the device will fail regardless. This option implies luks.
See cryptsetup(8) for possible values. The default is to try all key
slots in sequential order.
Added in version 209.
.RE
.PP keyfile-timeout=
Specifies the timeout for the device on which the key file resides or the
device used as the key file, and falls back to a password if it could not be
accessed. See systemd-cryptsetup-generator(8) for key files on
external devices.
Added in version 243.
.RE
.PP link-volume-key=
Specifies the kernel keyring and key description (see keyrings(7))
where LUKS2 volume key gets linked during device activation. The kernel
keyring description and key description must be separated by ::.
The kernel keyring part can be a string description or a predefined kernel
keyring prefixed with @ (e.g.: to use @s session or @u user keyring
directly). The type prefix text in the kernel keyring description is not
required. The specified kernel keyring must already exist at the time of
device activation.
The key part is a string description optionally prefixed by a %key_type:. If
no type is specified, the user type key is linked by default. See
keyctl(1) for more information on key descriptions (KEY IDENTIFIERS
section).
Note that the linked volume key is not cleaned up automatically when the
device is detached.
Added in version 256.
.RE
.PP luks
Force LUKS mode. When this mode is used, the following options are ignored
since they are provided by the LUKS header on the device: cipher=,
hash=, size=.
Added in version 186.
.RE
.PP bitlk
Decrypt BitLocker drive. Encryption parameters are deduced by cryptsetup from
BitLocker header.
Added in version 246.
.RE
.PP _netdev
Marks this cryptsetup device as requiring network. It will be started after
the network is available, similarly to systemd.mount(5) units marked
with _netdev. The service unit to set up this device will be ordered
between remote-fs-pre.target and remote-cryptsetup.target, instead of
cryptsetup-pre.target and cryptsetup.target.
Hint: if this device is used for a mount point that is specified in
fstab(5), the _netdev option should also be used for the mount
point. Otherwise, a dependency loop might be created where the mount point
will be pulled in by local-fs.target, while the service to configure the
network is usually only started after the local file system has been
mounted.
Added in version 235.
.RE
.PP noauto
This device will not be added to cryptsetup.target. This means that it will
not be automatically unlocked on boot, unless something else pulls it in. In
particular, if the device is used for a mount point, it'll be unlocked
automatically during boot, unless the mount point itself is also disabled
with noauto.
Added in version 186.
.RE
.PP nofail
This device will not be a hard dependency of cryptsetup.target. It'll still
be pulled in and started, but the system will not wait for the device to
show up and be unlocked, and boot will not fail if this is unsuccessful.
Note that other units that depend on the unlocked device may still fail. In
particular, if the device is used for a mount point, the mount point itself
also needs to have the nofail option, or the boot will fail if the
device is not unlocked successfully. If a keyfile and/or a header are
specified, the dependencies on their respective directories will also not be
fatal, so that unmounting said directories will not cause the generated
cryptset unit to be deactivated.
Added in version 186.
.RE
.PP offset=
Start offset in the backend device, in 512-byte sectors. This option is only
relevant for plain devices.
Added in version 220.
.RE
.PP plain
Force plain encryption mode.
Added in version 186.
.RE
.PP read-only, readonly
Set up the encrypted block device in read-only mode.
Added in version 186.
.RE
.PP same-cpu-crypt
Perform encryption using the same CPU that IO was submitted on. The default
is to use an unbound workqueue so that encryption work is automatically
balanced between available CPUs.
This requires kernel 4.0 or newer.
Added in version 242.
.RE
.PP submit-from-crypt-cpus
Disable offloading writes to a separate thread after encryption. There are
some situations where offloading write requests from the encryption threads
to a dedicated thread degrades performance significantly. The default is to
offload write requests to a dedicated thread because it benefits the CFQ
scheduler to have writes submitted using the same context.
This requires kernel 4.0 or newer.
Added in version 242.
.RE
.PP no-read-workqueue
Bypass dm-crypt internal workqueue and process read requests synchronously.
The default is to queue these requests and process them asynchronously.
This requires kernel 5.9 or newer.
Added in version 248.
.RE
.PP no-write-workqueue
Bypass dm-crypt internal workqueue and process write requests synchronously.
The default is to queue these requests and process them asynchronously.
This requires kernel 5.9 or newer.
Added in version 248.
.RE
.PP skip=
How many 512-byte sectors of the encrypted data to skip at the beginning.
This is different from the offset= option with respect to the sector
numbers used in initialization vector (IV) calculation. Using offset=
will shift the IV calculation by the same negative amount. Hence, if
offset=n is given, sector n will get a sector
number of 0 for the IV calculation. Using skip= causes sector
n to also be the first sector of the mapped device, but with its
number for IV generation being n.
This option is only relevant for plain devices.
Added in version 220.
.RE
.PP size=
Specifies the key size in bits. See cryptsetup(8) for possible values
and the default value of this option.
Added in version 186.
.RE
.PP sector-size=
Specifies the sector size in bytes. See cryptsetup(8) for possible
values and the default value of this option.
Added in version 240.
.RE
.PP swap
The encrypted block device will be used as a swap device, and will be
formatted accordingly after setting up the encrypted block device, with
mkswap(8). This option implies plain.
.if n .sp
Warning
Using the swap option will destroy the contents of the named partition
during every boot, so make sure the underlying block device is specified
correctly.
.sp .5v
Added in version 186.
.RE
.PP tcrypt
Use TrueCrypt encryption mode. When this mode is used, the following options
are ignored since they are provided by the TrueCrypt header on the device or
do not apply: cipher=, hash=, keyfile-offset=,
keyfile-size=, size=.
When this mode is used, the passphrase is read from the key file given in the
third field. Only the first line of this file is read, excluding the new
line character.
Note that the TrueCrypt format uses both passphrase and key files to derive a
password for the volume. Therefore, the passphrase and all key files need to
be provided. Use tcrypt-keyfile= to provide the absolute path to all
key files. When using an empty passphrase in combination with one or more
key files, use /dev/null as the password file in the third field.
Added in version 206.
.RE
.PP tcrypt-hidden
Use the hidden TrueCrypt volume. This option implies tcrypt.
This will map the hidden volume that is inside of the volume provided in the
second field. Please note that there is no protection for the hidden volume
if the outer volume is mounted instead. See cryptsetup(8) for more
information on this limitation.
Added in version 206.
.RE
.PP tcrypt-keyfile=
Specifies the absolute path to a key file to use for a TrueCrypt volume. This
implies tcrypt and can be used more than once to provide several key
files.
See the entry for tcrypt on the behavior of the passphrase and key
files when using TrueCrypt encryption mode.
Added in version 206.
.RE
.PP tcrypt-system
Use TrueCrypt in system encryption mode. This option implies
tcrypt.
Added in version 206.
.RE
.PP tcrypt-veracrypt
Check for a VeraCrypt volume. VeraCrypt is a fork of TrueCrypt that is mostly
compatible, but uses different, stronger key derivation algorithms that
cannot be detected without this flag. Enabling this option could
substantially slow down unlocking, because VeraCrypt's key derivation takes
much longer than TrueCrypt's. This option implies tcrypt.
Added in version 232.
.RE
.PP veracrypt-pim=
Specifies a custom Personal Iteration Multiplier (PIM) value, which can range
from 0..2147468 for standard veracrypt volumes and 0..65535 for veracrypt
system volumes. A value of 0 will imply the VeraCrypt default. This option
is only effective when tcrypt-veracrypt is set.
Note that VeraCrypt enforces a minimal allowed PIM value depending on the
password strength and the hash algorithm used for key derivation, however
veracrypt-pim= is not checked against these bounds. See Veracrypt
Personal Iterations Multiplier[1] documentation for more
information.
Added in version 254.
.RE
.PP timeout=
Specifies the timeout for querying for a password. If no unit is specified,
seconds is used. Supported units are s, ms, us, min, h, d. A timeout of 0
waits indefinitely (which is the default).
Added in version 186.
.RE
.PP tmp=
The encrypted block device will be prepared for using it as /tmp/; it will be
formatted using mkfs(8). Takes a file system type as argument, such
as ext4, xfs or btrfs. If no argument is specified defaults to ext4. This
option implies plain.
.if n .sp
Warning
Using the tmp option will destroy the contents of the named partition
during every boot, so make sure the underlying block device is specified
correctly.
.sp .5v
Added in version 186.
.RE
.PP tries=
Specifies the maximum number of times the user is queried for a password. The
default is 3. If set to 0, the user is queried for a password
indefinitely.
Added in version 186.
.RE
.PP headless=
Takes a boolean argument, defaults to false. If true, never query
interactively for the password/PIN. Useful for headless systems.
Added in version 249.
.RE
.PP verify
If the encryption password is read from console, it has to be entered twice
to prevent typos.
Added in version 186.
.RE
.PP password-echo=yes|no|masked
Controls whether to echo passwords or security token PINs that are read from
console. Takes a boolean or the special string masked. The default is
password-echo=masked.
If enabled, the typed characters are echoed literally. If disabled, the typed
characters are not echoed in any form, the user will not get feedback on
their input. If set to masked, an asterisk (*) is echoed for each character
typed. Regardless of which mode is chosen, if the user hits the tabulator
key (↹) at any time, or the backspace key (⌫) before any other
data has been entered, then echo is turned off.
Added in version 249.
.RE
.PP password-cache=yes|no|read-only
Controls whether to use cache for passwords or security token PINs. Takes a
boolean or the special string read-only. Defaults to yes.
If set to read-only, the kernel keyring is checked for a password/PIN before
requesting one interactively. If set to yes, in addition to checking the
keyring, any password/PIN entered interactively is cached in the keyring
with a 2.5-minute timeout before being purged.
Note that this option is not permitted for PKCS#11 security tokens. The
reasoning behind this is that PKCS#11 security tokens are usually configured
to lock after being supplied an invalid PIN multiple times, so using the
cache might inadvertently lock the token.
Added in version 257.
.RE
.PP pkcs11-uri=
Takes either the special value auto or an RFC7512 PKCS#11 URI[2]
pointing to a private key which is used to decrypt the encrypted key
specified in the third column of the line. This is useful for unlocking
encrypted volumes through PKCS#11 compatible security tokens or smartcards.
See below for an example how to set up this mechanism for unlocking a LUKS2
volume with a YubiKey security token.
If specified as auto the volume must be of type LUKS2 and must carry PKCS#11
security token metadata in its LUKS2 JSON token section. In this mode the
URI and the encrypted key are automatically read from the LUKS2 JSON token
header. Use systemd-cryptenroll(1) as a simple tool for enrolling
PKCS#11 security tokens or smartcards in a way compatible with auto. In this
mode the third column of the line should remain empty (that is, specified as
-).
The specified URI can refer directly to a private key stored on a token or
alternatively just to a slot or token, in which case a search for a suitable
private key will be performed. In this case, if multiple suitable objects
are found, the token is refused. The keyfile configured in the third column
of the line is used as is (i.e. in binary form, unprocessed). The resulting
decrypted key (for RSA) or derived shared secret (for ECC) is then Base64
encoded before it is used to unlock the LUKS volume.
Use systemd-cryptenroll --pkcs11-token-uri=list to list all suitable
PKCS#11 security tokens currently plugged in, along with their URIs.
Note that many newer security tokens that may be used as PKCS#11 security
token typically also implement the newer and simpler FIDO2 standard.
Consider using fido2-device= (described below) to enroll it via FIDO2
instead. Note that a security token enrolled via PKCS#11 cannot be used to
unlock the volume via FIDO2, unless also enrolled via FIDO2, and vice
versa.
Added in version 245.
.RE
.PP fido2-device=
Takes either the special value auto or the path to a hidraw device node (e.g.
/dev/hidraw1) referring to a FIDO2 security token that implements the
hmac-secret extension (most current hardware security tokens do). See below
for an example how to set up this mechanism for unlocking an encrypted
volume with a FIDO2 security token.
If specified as auto the FIDO2 token device is automatically discovered, as
it is plugged in.
FIDO2 volume unlocking requires a client ID hash (CID) to be configured via
fido2-cid= (see below) and a key to pass to the security token's HMAC
functionality (configured in the line's third column) to operate. If not
configured and the volume is of type LUKS2, the CID and the key are read
from LUKS2 JSON token metadata instead. Use systemd-cryptenroll(1) as
simple tool for enrolling FIDO2 security tokens for LUKS2 volumes.
Use systemd-cryptenroll --fido2-device=list to list all suitable FIDO2
security tokens currently plugged in, along with their device nodes.
This option implements the following mechanism: the configured key is hashed
via they HMAC keyed hash function the FIDO2 device implements, keyed by a
secret key embedded on the device. The resulting hash value is Base64
encoded and used to unlock the LUKS2 volume. As it should not be possible to
extract the secret from the hardware token, it should not be possible to
retrieve the hashed key given the configured key — without possessing
the hardware token.
Note that many security tokens that implement FIDO2 also implement PKCS#11,
suitable for unlocking volumes via the pkcs11-uri= option described
above. Typically the newer, simpler FIDO2 standard is preferable.
Added in version 248.
.RE
.PP fido2-cid=
Takes a Base64 encoded FIDO2 client ID to use for the FIDO2 unlock operation.
If specified, but fido2-device= is not, fido2-device=auto is
implied. If fido2-device= is used but fido2-cid= is not, the
volume must be of LUKS2 type, and the CID is read from the LUKS2 JSON token
header. Use systemd-cryptenroll(1) for enrolling a FIDO2 token in the
LUKS2 header compatible with this automatic mode.
Added in version 248.
.RE
.PP fido2-rp=
Takes a string, configuring the FIDO2 Relying Party (rp) for the FIDO2 unlock
operation. If not specified, io.systemd.cryptsetup is used, except if the
LUKS2 JSON token header contains a different value. It should normally not
be necessary to override this.
Added in version 248.
.RE
.PP fido2-pin=
Controls whether to require the user to enter a PIN when unlocking the volume
(the FIDO2 clientPin feature). This option only applies when in manual mode,
i.e. when fido2-cid= option is set. Defaults to neither true or
false, but rather to v248 behavior, that is: try with no PIN first,
but if token reports that PIN is required, try again asking for PIN.
Added in version 257.
.RE
.PP fido2-up=
Controls whether to require the user to verify presence (tap the token, the
FIDO2 up feature) when unlocking the volume. This option only applies when
in manual mode, i.e. when fido2-cid= option is set. Defaults to
neither true or false, but rather to v248 behavior, that is: try with
no UP first, but if token reports that UP is required, try again with UP
enabled.
Added in version 257.
.RE
.PP fido2-uv=
Controls whether to require user verification (the FIDO2 uv feature) when
unlocking the volume. This option only applies when in manual mode, i.e.
when fido2-cid= option is set. Defaults to neither true or false, but
rather to v248 behavior, that is: omit configuring UV whatsoever.
Added in version 257.
.RE
.PP tpm2-device=
Takes either the special value auto or the path to a device node (e.g.
/dev/tpmrm0) referring to a TPM2 security chip. See below for an example how
to set up this mechanism for unlocking an encrypted volume with a TPM2
chip.
Use tpm2-pcrs= (see below) to configure the set of TPM2 PCRs to bind
the volume unlocking to. Use systemd-cryptenroll(1) as simple tool
for enrolling TPM2 security chips in LUKS2 volumes.
If specified as auto the TPM2 device is automatically discovered. Use
systemd-cryptenroll --tpm2-device=list to list all suitable TPM2
devices currently available, along with their device nodes.
This option implements the following mechanism: when enrolling a TPM2 device
via systemd-cryptenroll on a LUKS2 volume, a randomized key unlocking
the volume is generated on the host and loaded into the TPM2 chip where it
is encrypted with an asymmetric "primary" key pair derived from
the TPM2's internal "seed" key. Neither the seed key nor the
primary key are permitted to ever leave the TPM2 chip — however, the
now encrypted randomized key may. It is saved in the LUKS2 volume JSON token
header. When unlocking the encrypted volume, the primary key pair is
generated on the TPM2 chip again (which works as long as the chip's seed key
is correctly maintained by the TPM2 chip), which is then used to decrypt (on
the TPM2 chip) the encrypted key from the LUKS2 volume JSON token header
saved there during enrollment. The resulting decrypted key is then used to
unlock the volume. When the randomized key is encrypted the current values
of the selected PCRs (see below) are included in the operation, so that
different PCR state results in different encrypted keys and the decrypted
key can only be recovered if the same PCR state is reproduced.
Added in version 248.
.RE
.PP tpm2-pcrs=
Takes a + separated list of numeric TPM2 PCR (i.e. "Platform
Configuration Register") indexes to bind the TPM2 volume unlocking to.
This option is only useful when TPM2 enrollment metadata is not available in
the LUKS2 JSON token header already, the way systemd-cryptenroll
writes it there. If not used (and no metadata in the LUKS2 JSON token header
defines it), defaults to a list of a single entry: PCR 7. Assign an empty
string to encode a policy that binds the key to no PCRs, making the key
accessible to local programs regardless of the current PCR state.
Added in version 248.
.RE
.PP tpm2-pin=
Takes a boolean argument, defaults to false. Controls whether TPM2 volume
unlocking is bound to a PIN in addition to PCRs. Similarly, this option is
only useful when TPM2 enrollment metadata is not available.
Added in version 251.
.RE
.PP tpm2-signature=
Takes an absolute path to a TPM2 PCR JSON signature file, as produced by the
systemd-measure(1) tool. This permits locking LUKS2 volumes to any
PCR values for which a valid signature matching a public key specified at
key enrollment time can be provided. See systemd-cryptenroll(1) for
details on enrolling TPM2 PCR public keys. If this option is not specified
but it is attempted to unlock a LUKS2 volume with a signed TPM2 PCR
enrollment, a suitable signature file tpm2-pcr-signature.json is searched
for in /etc/systemd/, /run/systemd/, /usr/lib/systemd/ (in this order).
Added in version 252.
.RE
.PP tpm2-pcrlock=
Takes an absolute path to a TPM2 pcrlock policy file, as produced by the
systemd-pcrlock(8) tool. This permits locking LUKS2 volumes to a
local policy of allowed PCR values with variants. See
systemd-cryptenroll(1) for details on enrolling TPM2 pcrlock
policies. If this option is not specified but it is attempted to unlock a
LUKS2 volume with a TPM2 pcrlock enrollment, a suitable signature file
pcrlock.json is searched for in /run/systemd/ and /var/lib/systemd/ (in this
order).
Added in version 255.
.RE
.PP tpm2-measure-pcr=
Controls whether to measure the volume key of the encrypted volume to a TPM2
PCR. If set to "no" (which is the default) no PCR extension is
done. If set to "yes" the volume key is measured into PCR 15. If
set to a decimal integer in the range 0...23 the volume key is measured into
the specified PCR. The volume key is measured along with the activated
volume name and its UUID. This functionality is particularly useful for the
encrypted volume backing the root file system, as it then allows later TPM
objects to be securely bound to the root file system and hence the specific
installation.
Added in version 253.
.RE
.PP tpm2-measure-bank=
This option is deprecated and has no effect since version 262. All suitable
PCR banks are used.
Added in version 253.
.RE
.PP tpm2-measure-keyslot-nvpcr=
Controls whether to measure information about the used LUKS unlock keyslot to
a TPM2 non-volatile index (nvindex in PCR mode). Takes a boolean argument,
or an NvPCR name. If set to false or an empty string (which is the default)
no TPM2 nvindex extension is done, otherwise keyslot information is measured
to an nvindex of the specified name, which is allocated if needed. If set to
true the recommended default of cryptsetup is selected as NvPCR. The slot
index and the used unlock mechanism (i.e. tpm2, fido2, pkcs11) are measured
along with the activated volume name and its UUID.
Added in version 259.
.RE
.PP token-timeout=
Specifies how long to wait at most for configured security devices (i.e.
FIDO2, PKCS#11, TPM2) to show up. Takes a time value in seconds (but other
time units may be specified too, see systemd.time(7) for supported
formats). Defaults to 30s. Once the specified timeout elapsed authentication
via password is attempted. Note that this timeout applies to waiting for the
security device to show up — it does not apply to the PIN prompt for
the device (should one be needed) or similar. Pass 0 to turn off the timeout
and wait forever.
Added in version 250.
.RE
.PP try-empty-password=
Takes a boolean argument. If enabled, right before asking the user for a
password it is first attempted to unlock the volume with an empty password.
This is useful for systems that are initialized with an encrypted volume
with only an empty password set, which shall be replaced with a suitable
password during first boot, but after activation.
Added in version 246.
.RE
.PP x-systemd.device-timeout=
Specifies how long systemd should wait for a block device to show up before
giving up on the entry. The argument is a time in seconds or explicitly
specified units of s, min, h, ms.
Added in version 216.
.RE
.PP x-initrd.attach
Setup this encrypted block device in the initrd, similarly to
systemd.mount(5) units marked with x-initrd.mount.
Although it is not necessary to mark the mount entry for the root file system
with x-initrd.mount, x-initrd.attach is still recommended with
the encrypted block device containing the root file system as otherwise
systemd will attempt to detach the device during the regular system shutdown
while it is still in use. With this option the device will still be detached
but later after the root file system is unmounted.
All other encrypted block devices that contain file systems mounted in the
initrd should use this option.
Added in version 245.
.RE
.PP fixate-volume-key=
Pin the expected hash of the volume key.
In certain cases, e.g. for LUKS volumes where the key is sealed to the TPM2,
this may be required to provide a guarantee that the volume being attached
is the volume which was previously created. fixate-volume-key= can be
used to set the expected volume key hash and refuse to attach the volume if
it has a different one. The expected hash matches the digest which is
measured to the sha256 PCR bank of the TPM2 when tpm2-measure-pcr= is
used.
For newly created LUKS volumes, the expected hash can be generated by
systemd-repart(8). For additional details, see the
EncryptedVolume= description at repart.d(5).
Added in version 260.
.RE
.PP At early boot and when the system manager configuration is reloaded, this
file is translated into native systemd units by
systemd-cryptsetup-generator(8).
.SH "AF_UNIX KEY FILES"
.PP If the key file path (as specified in the third column of /etc/crypttab
entries, see above) refers to an AF_UNIX stream socket in the file
system, the key is acquired by connecting to the socket and reading the key
from the connection. The connection is made from an AF_UNIX socket
name in the abstract namespace, see unix(7) for details. The source
socket name is chosen according to the following format:
NUL RANDOM /cryptsetup/ VOLUME
.PP In other words: a NUL byte (as required for abstract namespace
sockets), followed by a random string (consisting of alphanumeric characters
only), followed by the literal string /cryptsetup/, followed by the name of
the volume to acquire they key for. For example, for the volume myvol:
\0d7067f78d9827418/cryptsetup/myvol
.PP Services listening on the AF_UNIX stream socket may query the
source socket name with getpeername(2), and use this to determine
which key to send, allowing a single listening socket to serve keys for
multiple volumes. If the PKCS#11 logic is used (see above), the socket
source name is picked in similar fashion, except that the literal string
/cryptsetup-pkcs11/ is used. And similarly for FIDO2
(/cryptsetup-fido2-salt/) and TPM2 (/cryptsetup-tpm2/). A different path
component is used so that services providing key material know that the
secret key was not requested directly, but instead an encrypted key that
will be decrypted via the PKCS#11/FIDO2/TPM2 logic to acquire the final
secret key.
.SH "EXAMPLES"
.PP Example 1. /etc/crypttab example
.PP Set up four encrypted block devices. One using LUKS for normal storage,
another one for usage as a swap device and two TrueCrypt volumes. For the
fourth device, the option string is interpreted as two options
cipher=xchacha12,aes-adiantum-plain64, keyfile-timeout=10s.
luks UUID=2505567a-9e27-4efe-a4d5-15ad146c258b swap /dev/sda7 /dev/urandom swap truecrypt /dev/sda2 /etc/container_password tcrypt hidden /mnt/tc_hidden /dev/null tcrypt-hidden,tcrypt-keyfile=/etc/keyfile external /dev/sda3 keyfile:LABEL=keydev keyfile-timeout=10s,cipher=xchacha12\,aes-adiantum-plain64
.PP Example 2. Yubikey-based PKCS#11 Volume Unlocking
Example
.PP The PKCS#11 logic allows hooking up any compatible security token that is
capable of storing RSA or EC cryptographic keys for unlocking an encrypted
volume. Here's an example how to set up a Yubikey security token for this
purpose on a LUKS2 volume, using ykmap(1) from the yubikey-manager
project to initialize the token and systemd-cryptenroll(1) to add it
in the LUKS2 volume:
# SPDX-License-Identifier: MIT-0 # Destroy any old key on the Yubikey (careful!) ykman piv reset # Generate a new private/public key pair on the device, store the public key in # 'pubkey.pem'. ykman piv generate-key -a RSA2048 9d pubkey.pem # Create a self-signed certificate from this public key, and store it on the # device. The "subject" should be an arbitrary user-chosen string to identify # the token with. ykman piv generate-certificate --subject "Knobelei" 9d pubkey.pem # We do not need the public key anymore, let's remove it. Since it is not # security sensitive we just do a regular "rm" here. rm pubkey.pem # Enroll the freshly initialized security token in the LUKS2 volume. Replace # /dev/sdXn by the partition to use (e.g. /dev/sda1). sudo systemd-cryptenroll --pkcs11-token-uri=auto /dev/sdXn # Test: Let's run systemd-cryptsetup to test if this all worked. sudo systemd-cryptsetup attach mytest /dev/sdXn none pkcs11-uri=auto # If that worked, let's now add the same line persistently to /etc/crypttab, # for the future. We do not want to use the (unstable) /dev/sdX name, so let's # figure out a stable link: udevadm info -q symlink -r /dev/sdXn # Now add the line using the by-uuid symlink to /etc/crypttab: sudo bash -c 'echo "mytest /dev/disk/by-uuid/... none pkcs11-uri=auto" >>/etc/crypttab' # Depending on your distribution and encryption setup, you may need to manually # regenerate your initramfs to be able to use a Yubikey / PKCS#11 token to # unlock the partition during early boot. # More information at https://unix.stackexchange.com/a/705809. # On Fedora based systems: sudo dracut --force # On Debian based systems: sudo update-initramfs -u
.PP A few notes on the above:
.PP Example 3. FIDO2 Volume Unlocking Example
.PP The FIDO2 logic allows using any compatible FIDO2 security token that
implements the hmac-secret extension for unlocking an encrypted volume.
Here's an example how to set up a FIDO2 security token for this purpose for
a LUKS2 volume, using systemd-cryptenroll(1):
# SPDX-License-Identifier: MIT-0 # Enroll the security token in the LUKS2 volume. Replace /dev/sdXn by the # partition to use (e.g. /dev/sda1). sudo systemd-cryptenroll --fido2-device=auto /dev/sdXn # Test: Let's run systemd-cryptsetup to test if this worked. sudo systemd-cryptsetup attach mytest /dev/sdXn none fido2-device=auto # If that worked, let's now add the same line persistently to /etc/crypttab, # for the future. We do not want to use the (unstable) /dev/sdX name, so let's # figure out a stable link: udevadm info -q symlink -r /dev/sdXn # Now add the line using the by-uuid symlink to /etc/crypttab: sudo bash -c 'echo "mytest /dev/disk/by-uuid/... none fido2-device=auto" >>/etc/crypttab' # Depending on your distribution and encryption setup, you may need to manually # regenerate your initramfs to be able to use a FIDO2 device to unlock the # partition during early boot. # More information at https://unix.stackexchange.com/a/705809. # On Fedora based systems: sudo dracut --force # On Debian based systems: sudo update-initramfs -u
.PP Example 4. TPM2 Volume Unlocking Example
.PP The TPM2 logic allows using any TPM2 chip supported by the Linux kernel
for unlocking an encrypted volume. Here's an example how to set up a TPM2
chip for this purpose for a LUKS2 volume, using
systemd-cryptenroll(1):
# SPDX-License-Identifier: MIT-0 # Enroll the TPM2 security chip in the LUKS2 volume, and bind it to PCR 7 # only. Replace /dev/sdXn by the partition to use (e.g. /dev/sda1). sudo systemd-cryptenroll --tpm2-device=auto --tpm2-pcrs=7 /dev/sdXn # Test: Let's run systemd-cryptsetup to test if this worked. sudo systemd-cryptsetup attach mytest /dev/sdXn none tpm2-device=auto # If that worked, let's now add the same line persistently to /etc/crypttab, # for the future. We do not want to use the (unstable) /dev/sdX name, so let's # figure out a stable link: udevadm info -q symlink -r /dev/sdXn # Now add the line using the by-uuid symlink to /etc/crypttab: sudo bash -c 'echo "mytest /dev/disk/by-uuid/... none tpm2-device=auto" >>/etc/crypttab' # And now let's check that automatic unlocking works: sudo systemd-cryptsetup detach mytest sudo systemctl daemon-reload sudo systemctl start cryptsetup.target systemctl is-active systemd-cryptsetup@mytest.service # Once we have the device which will be unlocked automatically, we can use it. # Usually we would create a file system and add it to /etc/fstab: sudo mkfs.ext4 /dev/mapper/mytest # This prints a 'Filesystem UUID', which we can use as a stable name: sudo bash -c 'echo "/dev/disk/by-uuid/... /var/mytest ext4 defaults,x-systemd.mkdir 0 2" >>/etc/fstab' # And now let's check that the mounting works: sudo systemctl daemon-reload sudo systemctl start /var/mytest systemctl status /var/mytest # Depending on your distribution and encryption setup, you may need to manually # regenerate your initramfs to be able to use a TPM2 security chip to unlock # the partition during early boot. # More information at https://unix.stackexchange.com/a/705809. # On Fedora based systems: sudo dracut --force # On Debian based systems: sudo update-initramfs -u
.SH "SEE ALSO"
.PP systemd(1), systemd-cryptsetup@.service(8),
systemd-cryptsetup-generator(8), systemd-cryptenroll(1),
systemd-repart(8), repart.d(5), fstab(5),
cryptsetup(8), mkswap(8), mke2fs(8)
NOTES
- 1.
- Veracrypt Personal Iterations Multiplier
- 2.
- RFC7512 PKCS#11 URI
- 3.
- Yubico PIV certificate slots
| systemd 262.2 |