cryptsetup/man/integritysetup.8.adoc

= integritysetup(8)
:doctype: manpage
:manmanual: Maintenance Commands
:mansource: integritysetup {release-version}
:man-linkstyle: pass:[blue R < >]

== NAME

integritysetup - utility for configuring and managing dm-integrity devices

== SYNOPSIS

*integritysetup <action> [<options>] <action args>*

== DESCRIPTION

*Integritysetup* is a utility for configuring and managing kernel dm-integrity devices.

Kernel device-mapper *dm-integrity* target emulates an additional data integrity tag per disk sector and provides transparent data integrity protection of block devices.

You can configure these additional integrity tags directly with *integritysetup*, or indirectly (for authenticated encryption) through *LUKS2* and *cryptsetup*(8).
Unlike *dm-verity*, *dm-integrity* devices support both read and write operations.
The kernel performs data integrity checking transparently using a selected checksum or cryptographic hash algorithm.

Integrity devices can be activated during boot through *integritytab*(5), which is part of *systemd*(1).

== BASIC ACTIONS

Integritysetup supports these operations:

=== FORMAT
*format <device>*

Formats <device> (calculates space and dm-integrity superblock and wipes the device).

*<options>* can be [--data-device, --batch-mode, --no-wipe, --journal-size, --interleave-sectors, --tag-size, --integrity, --integrity-key-size, --integrity-key-file, --sector-size, --progress-frequency, --progress-json].

=== OPEN
*open <device> <name>* +
create <name> <device> (OBSOLETE syntax)

Open a mapping with <name> backed by device <device>.

If the integrity algorithm of the device is non-default, then the algorithm should be specified with the --integrity option.
This will not be detected from the device.

*<options>* can be [--data-device, --batch-mode, --journal-watermark, --journal-commit-time, --buffer-sectors, --integrity, --integrity-key-size, --integrity-key-file, --integrity-no-journal, --integrity-recalculate, --integrity-recalculate-reset,--integrity-recovery-mode, --allow-discards].

=== CLOSE
*close <name>* +
remove <name> (OBSOLETE syntax)

Removes existing mapping <name>.

*<options>* can be [--deferred] or [--cancel-deferred].

=== STATUS
*status <name>*

Reports status for the active integrity mapping <name>.

=== DUMP
*dump <device>*

Report parameters from the on-disk stored superblock.

=== RESIZE
*resize <name>*

Resizes an active mapping <name>.

If --size (in 512-byte sectors) or --device-size is not specified, the size is computed from the underlying device.
After resize, the *recalculating* flag is set.
If --wipe flag is set and the size of the device is increased, the newly added section will be wiped.

Increasing the size of integrity volumes has been possible since the Linux kernel version 5.7; shrinking should work on older kernels, too.

*<options>* can be [--size, --device-size, --wipe].

== OPTIONS
*--allow-discards*::
Allow the use of discard (TRIM) requests for the device.
This option is available since the Linux kernel version 5.7.

*--batch-mode*, *-q*::
Do not ask for confirmation.

*--bitmap-flush-time* _ms_::
Bitmap flush time in milliseconds.
+
*WARNING:*
In case of a crash, it is possible that the data and integrity tag don't match if the journal is disabled.

*--bitmap-sectors-per-bit* _sectors_::
The number of 512-byte sectors per bitmap bit must be a power of two.

*--buffer-sectors* _sectors_::
The number of sectors in one buffer.
+
The tag area is accessed using buffers; the large buffer size means the I/O size will be larger, but there could be less I/Os issued.

*--cancel-deferred*::
Removes a previously configured deferred device removal in the *close* command.

*--data-device* _<data_device>_::
Specify a separate data device that contains existing data.
The <device> will then contain calculated integrity tags and a journal for data on <data_device>.
+
To not wipe the data device after initial format, also specify --no-wipe option and activate with --integrity-recalculate to recalculate integrity tags automatically.

*--debug*::
Run in debug mode with full diagnostic logs.
Debug output lines are always prefixed by *#*.

*--deferred*::
Defers device removal in the *close* command until the last user closes it.

*--help*, *-?*::
Show help text and default parameters.

*--integrity*, *-I* _algorithm_::
Use internal integrity calculation (standalone mode).
The integrity algorithm can be CRC (crc32c/crc32), a non-cryptographic hash function (xxhash64) or a hash function (sha1, sha256).
+
For HMAC (hmac-sha256), you must specify an integrity key and its size.

*--integrity-bitmap-mode*, *-B*::
Use alternate bitmap mode (available since Linux kernel 5.2), where dm-integrity uses a bitmap instead of a journal.
If a bit in the bitmap is 1, the corresponding region's data and integrity tags are not synchronized - if the machine crashes, the unsynchronized regions will be recalculated.
The bitmap mode is faster than the journal mode because we don't have to write the data twice.
However, it is also less reliable because if data corruption happens when the machine crashes, it may not be detected.

*--integrity-inline*::
Store integrity tags in hardware sector integrity fields.
The device must support sectors with additional protection information (PI, also known as DIF - data integrity field) of the requested size.
Another storage subsystem must not use the additional field (the device must present a "nop" profile in the kernel).
Note that some devices must be reformatted at a low level to support this option; for NVMe devices, see nvme(1) id-ns LBA profiles.
+
No journal or bitmap is used in this mode.
The device should operate with native speed (without any overhead).
+
This option is available since the Linux kernel version 6.11.

*--integrity-key-file* _file_::
The file with the integrity key.

*--integrity-key-size* _bytes_::
The size of the data integrity key.
Maximum is 4096 bytes.

*--integrity-no-journal*, *-D*::
Disable the journal for the integrity device.

*--integrity-recalculate*::
Automatically recalculate integrity tags in the kernel on activation.
The device can be used during automatic integrity recalculation, but becomes fully integrity protected only after the background operation is finished.
+
The primary intended use case is to skip initialization (wiping) of the data device after the initial format (see --no-wipe option).
This parameter can be used for activation, then the kernel will recalculate integrity tags in the background.
The integrity superblock contains a device offset that indicates the boundary to which the integrity tags are already updated.
You can check this offset with the dump command.

*--integrity-recalculate-reset*::
Restart recalculation from the beginning of the device.
It can be used to change the integrity checksum function.
Note, it does not change the tag length.
This option is available since the Linux kernel version 5.13.

*--integrity-recovery-mode*, *-R*::
Recovery mode (no journal, no tag checking).

*--interleave-sectors* _sectors_::
The number of interleaved sectors.

*--journal-commit-time* _ms_::
Commit time in milliseconds.
The journal is written when this time passes (and no explicit flush operation was issued).

*--journal-crypt* _algorithm_::
Encryption algorithm for the journal data area.
You can use a block cipher here, such as cbc-aes or a stream cipher, for example, chacha20 or ctr-aes.
+
The journal encryption options are only intended for testing.
Using journal encryption does not make sense without encryption of the data.

*--journal-crypt-key-file* _file_::
The file with the journal encryption key.

*--journal-crypt-key-size* _bytes_::
The size of the journal encryption key.
Maximum is 4096 bytes.

*--journal-integrity* _algorithm_::
Integrity algorithm for the journal area.
See --integrity option for detailed specification.

*--journal-integrity-key-file* _file_::
The file with the integrity key.

*--journal-integrity-key-size* _bytes_::
The size of the journal integrity key.
Maximum is 4096 bytes.

*--journal-size*, *-j* _bytes_::
Size of the journal.

*--journal-watermark* _percent_::
Journal watermark in percent.
When the journal size exceeds this watermark, the journal flush will be started.

*--no-wipe*::
Do not wipe the device after formatting.
A device that is not initially wiped will contain invalid checksums.

*--progress-frequency* _seconds_::
Print a separate line every <seconds> with wipe progress.

*--progress-json*::
Prints wipe progress data in JSON format, which is suitable mostly for machine processing.
It prints a separate line every half second (or based on --progress-frequency value).
The JSON output looks as follows during wipe progress (except it's a compact single line):
+
....
{
  "device":"/dev/sda",      // backing device or file
  "device_bytes":"8192",    // bytes wiped so far
  "device_size":"44040192", // total bytes to wipe
  "speed":"126877696",      // calculated speed in bytes per second (based on progress so far)
  "eta_ms":"2520012",       // estimated time to finish wipe in milliseconds
  "time_ms":"5561235"       // total time spent wiping device in milliseconds
}
....
+
Note on numbers in JSON output: Due to JSON parsers' limitations, all numbers are represented in a string format due to the need for full 64-bit unsigned integers.

*--sector-size*, *-s* _bytes_::
Sector size (power of two: 512, 1024, 2048, 4096).

*--tag-size*, *-t* _bytes_::
Size of the integrity tag per-sector (here, the integrity function will store the authentication tag).
+
The size can be smaller than the output size of the hash function; in that case, only part of the hash will be stored.

*--usage*::
Show short option help.

*--verbose*, *-v*::
Print more information on command execution.

*--version*, *-V*::
Show the program version.

*--wipe*::
Wipe the newly allocated area after resizing to a bigger size.
If this flag is not set, checksums will be calculated for previously stored data in the newly allocated area.

== LEGACY COMPATIBILITY OPTIONS

Do not use these options until you need compatibility with a specific old kernel.

*--integrity-legacy-padding*::
Use inefficient legacy padding.

*--integrity-legacy-hmac*::
Use old flawed HMAC calculation (also does not protect superblock).

*--integrity-legacy-recalculate*::
Allow insecure recalculating of volumes with HMAC keys (recalculation offset in superblock is not protected).

== RETURN CODES

Integritysetup returns *0* on success and a non-zero value on error.

Error codes are: *1* wrong parameters, *2* no permission, *3* out of memory, *4* wrong device specified, *5* device already exists or device is busy.

== NOTES

Format and activation of an integrity device always require superuser privilege because the superblock is calculated and handled in the dm-integrity kernel target.

== EXAMPLES

Format the device with default standalone mode (CRC32C):

*integritysetup format <device>*

Open the device with default parameters:

*integritysetup open <device> test*

Format the device in standalone mode for use with HMAC(SHA256):

*integritysetup format <device> --tag-size 32 --integrity hmac-sha256 --integrity-key-file <keyfile> --integrity-key-size <key_bytes>*

Open (activate) the device with HMAC(SHA256) and HMAC key in file:

*integritysetup open <device> test --integrity hmac-sha256 --integrity-key-file <keyfile> --integrity-key-size <key_bytes>*

Dump dm-integrity superblock information:

*integritysetup dump <device>*

== DM-INTEGRITY ON-DISK FORMAT

The on-disk format specification is available on the https://gitlab.com/cryptsetup/cryptsetup/wikis/DMIntegrity[DMIntegrity] page.

== AUTHORS

The integritysetup tool is written by mailto:gmazyland@gmail.com[Milan Broz].

include::man/common_footer.adoc[]