docs/ENVIRONMENT.md

   1 ---
   2 title: Known Environment Variables
   3 category: Interfaces
   4 layout: default
   5 SPDX-License-Identifier: LGPL-2.1-or-later
   6 ---
   7
   8 # Known Environment Variables
   9
  10 A number of systemd components take additional runtime parameters via
  11 environment variables. Many of these environment variables are not supported at
  12 the same level as command line switches and other interfaces are: we don't
  13 document them in the man pages and we make no stability guarantees for
  14 them. While they generally are unlikely to be dropped any time soon again, we
  15 do not want to guarantee that they stay around for good either.
  16
  17 Below is an (incomprehensive) list of the environment variables understood by
  18 the various tools. Note that this list only covers environment variables not
  19 documented in the proper man pages.
  20
  21 All tools:
  22
  23 * `$SYSTEMD_OFFLINE=[0|1]` — if set to `1`, then `systemctl` will refrain from
  24   talking to PID 1; this has the same effect as the historical detection of
  25   `chroot()`. Setting this variable to `0` instead has a similar effect as
  26   `$SYSTEMD_IGNORE_CHROOT=1`; i.e. tools will try to communicate with PID 1
  27   even if a `chroot()` environment is detected. You almost certainly want to
  28   set this to `1` if you maintain a package build system or similar and are
  29   trying to use a modern container system and not plain `chroot()`.
  30
  31 * `$SYSTEMD_IGNORE_CHROOT=1` — if set, don't check whether being invoked in a
  32   `chroot()` environment. This is particularly relevant for systemctl, as it
  33   will not alter its behaviour for `chroot()` environments if set. Normally it
  34   refrains from talking to PID 1 in such a case; turning most operations such
  35   as `start` into no-ops.  If that's what's explicitly desired, you might
  36   consider setting `$SYSTEMD_OFFLINE=1`.
  37
  38 * `$SD_EVENT_PROFILE_DELAYS=1` — if set, the sd-event event loop implementation
  39   will print latency information at runtime.
  40
  41 * `$SYSTEMD_PROC_CMDLINE` — if set, the contents are used as the kernel command
  42   line instead of the actual one in `/proc/cmdline`. This is useful for
  43   debugging, in order to test generators and other code against specific kernel
  44   command lines.
  45
  46 * `$SYSTEMD_FSTAB` — if set, use this path instead of `/etc/fstab`. Only useful
  47   for debugging.
  48
  49 * `$SYSTEMD_CRYPTTAB` — if set, use this path instead of `/etc/crypttab`. Only
  50   useful for debugging. Currently only supported by
  51   `systemd-cryptsetup-generator`.
  52
  53 * `$SYSTEMD_INTEGRITYTAB` — if set, use this path instead of
  54   `/etc/integritytab`. Only useful for debugging. Currently only supported by
  55   `systemd-integritysetup-generator`.
  56
  57 * `$SYSTEMD_VERITYTAB` — if set, use this path instead of
  58   `/etc/veritytab`. Only useful for debugging. Currently only supported by
  59   `systemd-veritysetup-generator`.
  60
  61 * `$SYSTEMD_EFI_OPTIONS` — if set, used instead of the string in the
  62   `SystemdOptions` EFI variable. Analogous to `$SYSTEMD_PROC_CMDLINE`.
  63
  64 * `$SYSTEMD_DEFAULT_HOSTNAME` — override the compiled-in fallback hostname
  65   (relevant in particular for the system manager and `systemd-hostnamed`).
  66   Must be a valid hostname (either a single label or a FQDN).
  67
  68 * `$SYSTEMD_IN_INITRD=[auto|lenient|0|1]` — if set, specifies initrd detection
  69   method. Defaults to `auto`. Behavior is defined as follows:
  70   `auto`: Checks if `/etc/initrd-release` exists, and a temporary fs is mounted
  71           on `/`. If both conditions meet, then it's in initrd.
  72   `lenient`: Similar to `auto`, but the rootfs check is skipped.
  73   `0|1`: Simply overrides initrd detection. This is useful for debugging and
  74          testing initrd-only programs in the main system.
  75
  76 * `$SYSTEMD_BUS_TIMEOUT=SECS` — specifies the maximum time to wait for method call
  77   completion. If no time unit is specified, assumes seconds. The usual other units
  78   are understood, too (us, ms, s, min, h, d, w, month, y). If it is not set or set
  79   to 0, then the built-in default is used.
  80
  81 * `$SYSTEMD_MEMPOOL=0` — if set, the internal memory caching logic employed by
  82   hash tables is turned off, and libc `malloc()` is used for all allocations.
  83
  84 * `$SYSTEMD_EMOJI=0` — if set, tools such as `systemd-analyze security` will
  85   not output graphical smiley emojis, but ASCII alternatives instead. Note that
  86   this only controls use of Unicode emoji glyphs, and has no effect on other
  87   Unicode glyphs.
  88
  89 * `$RUNTIME_DIRECTORY` — various tools use this variable to locate the
  90   appropriate path under `/run/`. This variable is also set by the manager when
  91   `RuntimeDirectory=` is used, see systemd.exec(5).
  92
  93 * `$SYSTEMD_CRYPT_PREFIX` — if set configures the hash method prefix to use for
  94   UNIX `crypt()` when generating passwords. By default the system's "preferred
  95   method" is used, but this can be overridden with this environment variable.
  96   Takes a prefix such as `$6$` or `$y$`. (Note that this is only honoured on
  97   systems built with libxcrypt and is ignored on systems using glibc's
  98   original, internal `crypt()` implementation.)
  99
 100 * `$SYSTEMD_RDRAND=0` — if set, the RDRAND instruction will never be used,
 101   even if the CPU supports it.
 102
 103 * `$SYSTEMD_SECCOMP=0` — if set, seccomp filters will not be enforced, even if
 104   support for it is compiled in and available in the kernel.
 105
 106 * `$SYSTEMD_LOG_SECCOMP=1` — if set, system calls blocked by seccomp filtering,
 107   for example in `systemd-nspawn`, will be logged to the audit log, if the
 108   kernel supports this.
 109
 110 `systemctl`:
 111
 112 * `$SYSTEMCTL_FORCE_BUS=1` — if set, do not connect to PID1's private D-Bus
 113   listener, and instead always connect through the dbus-daemon D-bus broker.
 114
 115 * `$SYSTEMCTL_INSTALL_CLIENT_SIDE=1` — if set, enable or disable unit files on
 116   the client side, instead of asking PID 1 to do this.
 117
 118 * `$SYSTEMCTL_SKIP_SYSV=1` — if set, do not call SysV compatibility hooks.
 119
 120 `systemd-nspawn`:
 121
 122 * `$SYSTEMD_NSPAWN_UNIFIED_HIERARCHY=1` — if set, force `systemd-nspawn` into
 123   unified cgroup hierarchy mode.
 124
 125 * `$SYSTEMD_NSPAWN_API_VFS_WRITABLE=1` — if set, make `/sys/`, `/proc/sys/`,
 126   and friends writable in the container. If set to "network", leave only
 127   `/proc/sys/net/` writable.
 128
 129 * `$SYSTEMD_NSPAWN_CONTAINER_SERVICE=…` — override the "service" name nspawn
 130   uses to register with machined. If unset defaults to "nspawn", but with this
 131   variable may be set to any other value.
 132
 133 * `$SYSTEMD_NSPAWN_USE_CGNS=0` — if set, do not use cgroup namespacing, even if
 134   it is available.
 135
 136 * `$SYSTEMD_NSPAWN_LOCK=0` — if set, do not lock container images when running.
 137
 138 * `$SYSTEMD_NSPAWN_TMPFS_TMP=0` — if set, do not overmount `/tmp/` in the
 139   container with a tmpfs, but leave the directory from the image in place.
 140
 141 * `$SYSTEMD_SUPPRESS_SYNC=1` — if set, all disk synchronization syscalls are
 142   blocked to the container payload (e.g. `sync()`, `fsync()`, `syncfs()`, …)
 143   and the `O_SYNC`/`O_DSYNC` flags are made unavailable to `open()` and
 144   friends. This is equivalent to passing `--suppress-sync=yes` on the
 145   `systemd-nspawn` command line.
 146
 147 `systemd-logind`:
 148
 149 * `$SYSTEMD_BYPASS_HIBERNATION_MEMORY_CHECK=1` — if set, report that
 150   hibernation is available even if the swap devices do not provide enough room
 151   for it.
 152
 153 * `$SYSTEMD_REBOOT_TO_FIRMWARE_SETUP` — if set, overrides `systemd-logind`'s
 154   built-in EFI logic of requesting a reboot into the firmware. Takes a boolean.
 155   If set to false, the functionality is turned off entirely. If set to true,
 156   instead of requesting a reboot into the firmware setup UI through EFI a file,
 157   `/run/systemd/reboot-to-firmware-setup` is created whenever this is
 158   requested. This file may be checked for by services run during system
 159   shutdown in order to request the appropriate operation from the firmware in
 160   an alternative fashion.
 161
 162 * `$SYSTEMD_REBOOT_TO_BOOT_LOADER_MENU` — similar to the above, allows
 163   overriding of `systemd-logind`'s built-in EFI logic of requesting a reboot
 164   into the boot loader menu. Takes a boolean. If set to false, the
 165   functionality is turned off entirely. If set to true, instead of requesting a
 166   reboot into the boot loader menu through EFI, the file
 167   `/run/systemd/reboot-to-boot-loader-menu` is created whenever this is
 168   requested. The file contains the requested boot loader menu timeout in µs,
 169   formatted in ASCII decimals, or zero in case no timeout is requested. This
 170   file may be checked for by services run during system shutdown in order to
 171   request the appropriate operation from the boot loader in an alternative
 172   fashion.
 173
 174 * `$SYSTEMD_REBOOT_TO_BOOT_LOADER_ENTRY` — similar to the above, allows
 175   overriding of `systemd-logind`'s built-in EFI logic of requesting a reboot
 176   into a specific boot loader entry. Takes a boolean. If set to false, the
 177   functionality is turned off entirely. If set to true, instead of requesting a
 178   reboot into a specific boot loader entry through EFI, the file
 179   `/run/systemd/reboot-to-boot-loader-entry` is created whenever this is
 180   requested. The file contains the requested boot loader entry identifier. This
 181   file may be checked for by services run during system shutdown in order to
 182   request the appropriate operation from the boot loader in an alternative
 183   fashion. Note that by default only boot loader entries which follow the [Boot
 184   Loader Specification](https://systemd.io/BOOT_LOADER_SPECIFICATION) and are
 185   placed in the ESP or the Extended Boot Loader partition may be selected this
 186   way. However, if a directory `/run/boot-loader-entries/` exists, the entries
 187   are loaded from there instead. The directory should contain the usual
 188   directory hierarchy mandated by the Boot Loader Specification, i.e. the entry
 189   drop-ins should be placed in
 190   `/run/boot-loader-entries/loader/entries/*.conf`, and the files referenced by
 191   the drop-ins (including the kernels and initrds) somewhere else below
 192   `/run/boot-loader-entries/`. Note that all these files may be (and are
 193   supposed to be) symlinks. `systemd-logind` will load these files on-demand,
 194   these files can hence be updated (ideally atomically) whenever the boot
 195   loader configuration changes. A foreign boot loader installer script should
 196   hence synthesize drop-in snippets and symlinks for all boot entries at boot
 197   or whenever they change if it wants to integrate with `systemd-logind`'s
 198   APIs.
 199
 200 `systemd-udevd`:
 201
 202 * `$NET_NAMING_SCHEME=` — if set, takes a network naming scheme (i.e. one of
 203   "v238", "v239", "v240"…, or the special value "latest") as parameter. If
 204   specified udev's `net_id` builtin will follow the specified naming scheme
 205   when determining stable network interface names. This may be used to revert
 206   to naming schemes of older udev versions, in order to provide more stable
 207   naming across updates. This environment variable takes precedence over the
 208   kernel command line option `net.naming-scheme=`, except if the value is
 209   prefixed with `:` in which case the kernel command line option takes
 210   precedence, if it is specified as well.
 211
 212 `nss-systemd`:
 213
 214 * `$SYSTEMD_NSS_BYPASS_SYNTHETIC=1` — if set, `nss-systemd` won't synthesize
 215   user/group records for the `root` and `nobody` users if they are missing from
 216   `/etc/passwd`.
 217
 218 * `$SYSTEMD_NSS_DYNAMIC_BYPASS=1` — if set, `nss-systemd` won't return
 219   user/group records for dynamically registered service users (i.e. users
 220   registered through `DynamicUser=1`).
 221
 222 * `$SYSTEMD_NSS_BYPASS_BUS=1` — if set, `nss-systemd` won't use D-Bus to do
 223   dynamic user lookups. This is primarily useful to make `nss-systemd` work
 224   safely from within `dbus-daemon`.
 225
 226 `systemd-timedated`:
 227
 228 * `$SYSTEMD_TIMEDATED_NTP_SERVICES=…` — colon-separated list of unit names of
 229   NTP client services. If set, `timedatectl set-ntp on` enables and starts the
 230   first existing unit listed in the environment variable, and
 231   `timedatectl set-ntp off` disables and stops all listed units.
 232
 233 `systemd-sulogin-shell`:
 234
 235 * `$SYSTEMD_SULOGIN_FORCE=1` — This skips asking for the root password if the
 236   root password is not available (such as when the root account is locked).
 237   See `sulogin(8)` for more details.
 238
 239 `bootctl` and other tools that access the EFI System Partition (ESP):
 240
 241 * `$SYSTEMD_RELAX_ESP_CHECKS=1` — if set, the ESP validation checks are
 242   relaxed. Specifically, validation checks that ensure the specified ESP path
 243   is a FAT file system are turned off, as are checks that the path is located
 244   on a GPT partition with the correct type UUID.
 245
 246 * `$SYSTEMD_ESP_PATH=…` — override the path to the EFI System Partition. This
 247   may be used to override ESP path auto detection, and redirect any accesses to
 248   the ESP to the specified directory. Note that unlike with `bootctl`'s
 249   `--path=` switch only very superficial validation of the specified path is
 250   done when this environment variable is used.
 251
 252 `systemd` itself:
 253
 254 * `$SYSTEMD_ACTIVATION_UNIT` — set for all NSS and PAM module invocations that
 255   are done by the service manager on behalf of a specific unit, in child
 256   processes that are later (after execve()) going to become unit
 257   processes. Contains the full unit name (e.g. "foobar.service"). NSS and PAM
 258   modules can use this information to determine in which context and on whose
 259   behalf they are being called, which may be useful to avoid deadlocks, for
 260   example to bypass IPC calls to the very service that is about to be
 261   started. Note that NSS and PAM modules should be careful to only rely on this
 262   data when invoked privileged, or possibly only when getppid() returns 1, as
 263   setting environment variables is of course possible in any even unprivileged
 264   contexts.
 265
 266 * `$SYSTEMD_ACTIVATION_SCOPE` — closely related to `$SYSTEMD_ACTIVATION_UNIT`,
 267   it is either set to `system` or `user` depending on whether the NSS/PAM
 268   module is called by systemd in `--system` or `--user` mode.
 269
 270 `systemd-remount-fs`:
 271
 272 * `$SYSTEMD_REMOUNT_ROOT_RW=1` — if set and no entry for the root directory
 273   exists in `/etc/fstab` (this file always takes precedence), then the root
 274   directory is remounted writable. This is primarily used by
 275   `systemd-gpt-auto-generator` to ensure the root partition is mounted writable
 276   in accordance to the GPT partition flags.
 277
 278 `systemd-firstboot` and `localectl`:
 279
 280 * `$SYSTEMD_LIST_NON_UTF8_LOCALES=1` — if set, non-UTF-8 locales are listed among
 281   the installed ones. By default non-UTF-8 locales are suppressed from the
 282   selection, since we are living in the 21st century.
 283
 284 `systemd-sysext`:
 285
 286 * `$SYSTEMD_SYSEXT_HIERARCHIES` — this variable may be used to override which
 287   hierarchies are managed by `systemd-sysext`. By default only `/usr/` and
 288   `/opt/` are managed, and directories may be added or removed to that list by
 289   setting this environment variable to a colon-separated list of absolute
 290   paths. Only "real" file systems and directories that only contain "real" file
 291   systems as submounts should be used. Do not specify API file systems such as
 292   `/proc/` or `/sys/` here, or hierarchies that have them as submounts. In
 293   particular, do not specify the root directory `/` here.
 294
 295 `systemd-tmpfiles`:
 296
 297 * `$SYSTEMD_TMPFILES_FORCE_SUBVOL` — if unset, `v`/`q`/`Q` lines will create
 298   subvolumes only if the OS itself is installed into a subvolume. If set to `1`
 299   (or another value interpreted as true), these lines will always create
 300   subvolumes if the backing filesystem supports them. If set to `0`, these
 301   lines will always create directories.
 302
 303 `systemd-sysv-generator`:
 304
 305 * `$SYSTEMD_SYSVINIT_PATH` — Controls where `systemd-sysv-generator` looks for
 306   SysV init scripts.
 307
 308 * `$SYSTEMD_SYSVRCND_PATH` — Controls where `systemd-sysv-generator` looks for
 309   SysV init script runlevel link farms.
 310
 311 systemd tests:
 312
 313 * `$SYSTEMD_TEST_DATA` — override the location of test data. This is useful if
 314   a test executable is moved to an arbitrary location.
 315
 316 * `$SYSTEMD_TEST_NSS_BUFSIZE` — size of scratch buffers for "reentrant"
 317   functions exported by the nss modules.
 318
 319 fuzzers:
 320
 321 * `$SYSTEMD_FUZZ_OUTPUT` — A boolean that specifies whether to write output to
 322   stdout. Setting to true is useful in manual invocations, since all output is
 323   suppressed by default.
 324
 325 * `$SYSTEMD_FUZZ_RUNS` — The number of times execution should be repeated in
 326   manual invocations.
 327
 328 Note that it may be also useful to set `$SYSTEMD_LOG_LEVEL`, since all logging
 329 is suppressed by default.
 330
 331 `systemd-importd`:
 332
 333 * `$SYSTEMD_IMPORT_BTRFS_SUBVOL` — takes a boolean, which controls whether to
 334   prefer creating btrfs subvolumes over plain directories for machine
 335   images. Has no effect on non-btrfs file systems where subvolumes are not
 336   available anyway. If not set, defaults to true.
 337
 338 * `$SYSTEMD_IMPORT_BTRFS_QUOTA` — takes a boolean, which controls whether to set
 339   up quota automatically for created btrfs subvolumes for machine images. If
 340   not set, defaults to true. Has no effect if machines are placed in regular
 341   directories, because btrfs subvolumes are not supported or disabled. If
 342   enabled, the quota group of the subvolume is automatically added to a
 343   combined quota group for all such machine subvolumes.
 344
 345 * `$SYSTEMD_IMPORT_SYNC` — takes a boolean, which controls whether to
 346   synchronize images to disk after installing them, before completing the
 347   operation. If not set, defaults to true. If disabled installation of images
 348   will be quicker, but not as safe.
 349
 350 `systemd-dissect`, `systemd-nspawn` and all other tools that may operate on
 351 disk images with `--image=` or similar:
 352
 353 * `$SYSTEMD_DISSECT_VERITY_SIDECAR` — takes a boolean, which controls whether to
 354   load "sidecar" Verity metadata files. If enabled (which is the default),
 355   whenever a disk image is used, a set of files with the `.roothash`,
 356   `.usrhash`, `.roothash.p7s`, `.usrhash.p7s`, `.verity` suffixes are searched
 357   adjacent to disk image file, containing the Verity root hashes, their
 358   signatures or the Verity data itself. If disabled this automatic discovery of
 359   Verity metadata files is turned off.
 360
 361 * `$SYSTEMD_DISSECT_VERITY_EMBEDDED` — takes a boolean, which controls whether
 362   to load the embedded Verity signature data. If enabled (which is the
 363   default), Verity root hash information and a suitable signature is
 364   automatically acquired from a signature partition, following the
 365   [Discoverable Partitions
 366   Specification](https://systemd.io/DISCOVERABLE_PARTITIONS). If disabled any
 367   such partition is ignored. Note that this only disables discovery of the root
 368   hash and its signature, the Verity data partition itself is still searched in
 369   the GPT image.
 370
 371 * `$SYSTEMD_DISSECT_VERITY_SIGNATURE` — takes a boolean, which controls whether
 372   to validate the signature of the Verity root hash if available. If enabled
 373   (which is the default), the signature of suitable disk images is validated
 374   against any of the certificates in `/etc/verity.d/*.crt` (and similar
 375   directories in `/usr/lib/`, `/run`, …) or passed to the kernel for validation
 376   against its built-in certificates.
 377
 378 * `$SYSTEMD_LOOP_DIRECT_IO` – takes a boolean, which controls whether to enable
 379   LO_FLAGS_DIRECT_IO (i.e. direct IO + asynchronous IO) on loopback block
 380   devices when opening them. Defaults to on, set this to "0" to disable this
 381   feature.
 382
 383 `systemd-cryptsetup`:
 384
 385 * `$SYSTEMD_CRYPTSETUP_USE_TOKEN_MODULE` – takes a boolean, which controls
 386   whether to use the libcryptsetup "token" plugin module logic even when
 387   activating via FIDO2, PKCS#11, TPM2, i.e. mechanisms natively supported by
 388   `systemd-cryptsetup`. Defaults to enabled.
 389
 390 Various tools that read passwords from the TTY, such as `systemd-cryptenroll`
 391 and `homectl`:
 392
 393 * `$PASSWORD` — takes a string: the literal password to use. If this
 394   environment variable is set it is used as password instead of prompting the
 395   user interactively. This exists primarily for debugging and testing
 396   purposes. Do not use this for production code paths, since environment
 397   variables are typically inherited down the process tree without restrictions
 398   and should thus not be used for secrets.
 399
 400 * `$NEWPASSWORD` — similar to `$PASSWORD` above, but is used when both a
 401   current and a future password are required, for example if the password is to
 402   be changed. In that case `$PASSWORD` shall carry the current (i.e. old)
 403   password and `$NEWPASSWORD` the new.
 404
 405 `systemd-homed`:
 406
 407 * `$SYSTEMD_HOME_ROOT` – defines an absolute path where to look for home
 408   directories/images. When unspecified defaults to `/home/`. This is useful for
 409   debugging purposes in order to run a secondary `systemd-homed` instance that
 410   operates on a different directory where home directories/images are placed.
 411
 412 * `$SYSTEMD_HOME_RECORD_DIR` – defines an absolute path where to look for
 413   fixated home records kept on the host. When unspecified defaults to
 414   `/var/lib/systemd/home/`. Similar to `$SYSTEMD_HOME_ROOT` this is useful for
 415   debugging purposes, in order to run a secondary `systemd-homed` instance that
 416   operates on a record database entirely separate from the host's.
 417
 418 * `$SYSTEMD_HOME_DEBUG_SUFFIX` – takes a short string that is suffixed to
 419   `systemd-homed`'s D-Bus and Varlink service names/sockets. This is also
 420   understood by `homectl`. This too is useful for running an additional copy of
 421   `systemd-homed` that doesn't interfere with the host's main one.
 422
 423 * `$SYSTEMD_HOMEWORK_PATH` – configures the path to the `systemd-homework`
 424   binary to invoke. If not specified defaults to
 425   `/usr/lib/systemd/systemd-homework`.
 426
 427   Combining these four environment variables is pretty useful when
 428   debugging/developing `systemd-homed`:
 429 ```sh
 430 SYSTEMD_HOME_DEBUG_SUFFIX=foo \
 431       SYSTEMD_HOMEWORK_PATH=/home/lennart/projects/systemd/build/systemd-homework \
 432       SYSTEMD_HOME_ROOT=/home.foo/ \
 433       SYSTEMD_HOME_RECORD_DIR=/var/lib/systemd/home.foo/ \
 434       /home/lennart/projects/systemd/build/systemd-homed
 435 ```
 436
 437 * `$SYSTEMD_HOME_MOUNT_OPTIONS_BTRFS`, `$SYSTEMD_HOME_MOUNT_OPTIONS_EXT4`,
 438   `$SYSTEMD_HOME_MOUNT_OPTIONS_XFS` – configure the default mount options to
 439   use for LUKS home directories, overriding the built-in default mount
 440   options. There's one variable for each of the supported file systems for the
 441   LUKS home directory backend.