docs/ENVIRONMENT.md

   1 ---
   2 title: Known Environment Variables
   3 category: Interfaces
   4 layout: default
   5 ---
   6
   7 # Known Environment Variables
   8
   9 A number of systemd components take additional runtime parameters via
  10 environment variables. Many of these environment variables are not supported at
  11 the same level as command line switches and other interfaces are: we don't
  12 document them in the man pages and we make no stability guarantees for
  13 them. While they generally are unlikely to be dropped any time soon again, we
  14 do not want to guarantee that they stay around for good either.
  15
  16 Below is an (incomprehensive) list of the environment variables understood by
  17 the various tools. Note that this list only covers environment variables not
  18 documented in the proper man pages.
  19
  20 All tools:
  21
  22 * `$SYSTEMD_OFFLINE=[0|1]` — if set to `1`, then `systemctl` will
  23   refrain from talking to PID 1; this has the same effect as the historical
  24   detection of `chroot()`.  Setting this variable to `0` instead has a similar
  25   effect as `SYSTEMD_IGNORE_CHROOT=1`; i.e. tools will try to
  26   communicate with PID 1 even if a `chroot()` environment is detected.
  27   You almost certainly want to set this to `1` if you maintain a package build system
  28   or similar and are trying to use a modern container system and not plain
  29   `chroot()`.
  30
  31 * `$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 systemd-cryptsetup-generator.
  51
  52 * `$SYSTEMD_EFI_OPTIONS` — if set, used instead of the string in the
  53   SystemdOptions EFI variable. Analogous to `$SYSTEMD_PROC_CMDLINE`.
  54
  55 * `$SYSTEMD_IN_INITRD` — takes a boolean. If set, overrides initrd detection.
  56   This is useful for debugging and testing initrd-only programs in the main
  57   system.
  58
  59 * `$SYSTEMD_BUS_TIMEOUT=SECS` — specifies the maximum time to wait for method call
  60   completion. If no time unit is specified, assumes seconds. The usual other units
  61   are understood, too (us, ms, s, min, h, d, w, month, y). If it is not set or set
  62   to 0, then the built-in default is used.
  63
  64 * `$SYSTEMD_MEMPOOL=0` — if set, the internal memory caching logic employed by
  65   hash tables is turned off, and libc malloc() is used for all allocations.
  66
  67 * `$SYSTEMD_EMOJI=0` — if set, tools such as "systemd-analyze security" will
  68   not output graphical smiley emojis, but ASCII alternatives instead. Note that
  69   this only controls use of Unicode emoji glyphs, and has no effect on other
  70   Unicode glyphs.
  71
  72 * `$RUNTIME_DIRECTORY` — various tools use this variable to locate the
  73   appropriate path under /run. This variable is also set by the manager when
  74   RuntimeDirectory= is used, see systemd.exec(5).
  75
  76 * `$SYSTEMD_CRYPT_PREFIX` — if set configures the hash method prefix to use for
  77   UNIX crypt() when generating passwords. By default the system's "preferred
  78   method" is used, but this can be overridden with this environment
  79   variable. Takes a prefix such as `$6$` or `$y$`. (Note that this is only
  80   honoured on systems built with libxcrypt and is ignored on systems using
  81   glibc's original, internal crypt() implementation.)
  82
  83 * `$SYSTEMD_RDRAND=0` — if set, the RDRAND instruction will never be used,
  84   even if the CPU supports it.
  85
  86 systemctl:
  87
  88 * `$SYSTEMCTL_FORCE_BUS=1` — if set, do not connect to PID1's private D-Bus
  89   listener, and instead always connect through the dbus-daemon D-bus broker.
  90
  91 * `$SYSTEMCTL_INSTALL_CLIENT_SIDE=1` — if set, enable or disable unit files on
  92   the client side, instead of asking PID 1 to do this.
  93
  94 * `$SYSTEMCTL_SKIP_SYSV=1` — if set, do not call out to SysV compatibility hooks.
  95
  96 * `$SYSTEMD_LOG_SECCOMP=1` — if set, system calls blocked by seccomp filtering,
  97   for example in systemd-nspawn, will be logged to the audit log, if the current
  98   kernel version supports this.
  99
 100 systemd-nspawn:
 101
 102 * `$SYSTEMD_NSPAWN_UNIFIED_HIERARCHY=1` — if set, force nspawn into unified
 103   cgroup hierarchy mode.
 104
 105 * `$SYSTEMD_NSPAWN_API_VFS_WRITABLE=1` — if set, make /sys and /proc/sys and
 106   friends writable in the container. If set to "network", leave only
 107   /proc/sys/net writable.
 108
 109 * `$SYSTEMD_NSPAWN_CONTAINER_SERVICE=…` — override the "service" name nspawn
 110   uses to register with machined. If unset defaults to "nspawn", but with this
 111   variable may be set to any other value.
 112
 113 * `$SYSTEMD_NSPAWN_USE_CGNS=0` — if set, do not use cgroup namespacing, even if
 114   it is available.
 115
 116 * `$SYSTEMD_NSPAWN_LOCK=0` — if set, do not lock container images when running.
 117
 118 * `$SYSTEMD_NSPAWN_TMPFS_TMP=0` — if set, do not overmount /tmp in the
 119   container with a tmpfs, but leave the directory from the image in place.
 120
 121 systemd-logind:
 122
 123 * `$SYSTEMD_BYPASS_HIBERNATION_MEMORY_CHECK=1` — if set, report that
 124   hibernation is available even if the swap devices do not provide enough room
 125   for it.
 126
 127 systemd-udevd:
 128
 129 * `$NET_NAMING_SCHEME=` – if set, takes a network naming scheme (i.e. one of
 130   "v238", "v239", "v240"…, or the special value "latest") as parameter. If
 131   specified udev's net_id builtin will follow the specified naming scheme when
 132   determining stable network interface names. This may be used to revert to
 133   naming schemes of older udev versions, in order to provide more stable naming
 134   across updates. This environment variable takes precedence over the kernel
 135   command line option `net.naming-scheme=`, except if the value is prefixed
 136   with `:` in which case the kernel command line option takes precedence, if it
 137   is specified as well.
 138
 139 * `$SYSTEMD_REBOOT_TO_FIRMWARE_SETUP` — if set overrides systemd-logind's
 140   built-in EFI logic of requesting a reboot into the firmware. Takes a
 141   boolean. If set to false the functionality is turned off entirely. If set to
 142   true instead of requesting a reboot into the firmware setup UI through EFI a
 143   file `/run/systemd/reboot-to-firmware-setup` is created whenever this is
 144   requested. This file may be checked for by services run during system
 145   shutdown in order to request the appropriate operation from the firmware in
 146   an alternative fashion.
 147
 148 * `$SYSTEMD_REBOOT_TO_BOOT_LOADER_MENU` — similar to the above, allows
 149   overriding of systemd-logind's built-in EFI logic of requesting a reboot into
 150   the boot loader menu. Takes a boolean. If set to false the functionality is
 151   turned off entirely. If set to true instead of requesting a reboot into the
 152   boot loader menu through EFI a file `/run/systemd/reboot-to-boot-loader-menu`
 153   is created whenever this is requested. The file contains the requested boot
 154   loader menu timeout in µs, formatted in ASCII decimals, or zero in case no
 155   timeout is requested. This file may be checked for by services run during
 156   system shutdown in order to request the appropriate operation from the boot
 157   loader in an alternative fashion.
 158
 159 * `$SYSTEMD_REBOOT_TO_BOOT_LOADER_ENTRY` — similar to the above, allows
 160   overriding of systemd-logind's built-in EFI logic of requesting a reboot into
 161   a specific boot loader entry. Takes a boolean. If set to false the
 162   functionality is turned off entirely. If set to true instead of requesting a
 163   reboot into a specific boot loader entry through EFI a file
 164   `/run/systemd/reboot-to-boot-loader-entry` is created whenever this is
 165   requested. The file contains the requested boot loader entry identifier. This
 166   file may be checked for by services run during system shutdown in order to
 167   request the appropriate operation from the boot loader in an alternative
 168   fashion. Note that by default only boot loader entries which follow the [Boot
 169   Loader Specification](https://systemd.io/BOOT_LOADER_SPECIFICATION) and are
 170   placed in the ESP or the Extended Boot Loader partition may be selected this
 171   way. However, if a directory `/run/boot-loader-entries/` exists, the entries
 172   are loaded from there instead. The directory should contain the usual
 173   directory hierarchy mandated by the Boot Loader Specification, i.e. the entry
 174   drop-ins should be placed in
 175   `/run/boot-loader-entries/loader/entries/*.conf`, and the files referenced by
 176   the drop-ins (including the kernels and initrds) somewhere else below
 177   `/run/boot-loader-entries/`. Note that all these files may be (and are
 178   supposed to be) symlinks. systemd-logind will load these files on-demand,
 179   these files can hence be updated (ideally atomically) whenever the boot
 180   loader configuration changes. A foreign boot loader installer script should
 181   hence synthesize drop-in snippets and symlinks for all boot entries at boot
 182   or whenever they change if it wants to integrate with systemd-logind's APIs.
 183
 184 installed systemd tests:
 185
 186 * `$SYSTEMD_TEST_DATA` — override the location of test data. This is useful if
 187   a test executable is moved to an arbitrary location.
 188
 189 nss-systemd:
 190
 191 * `$SYSTEMD_NSS_BYPASS_SYNTHETIC=1` — if set, `nss-systemd` won't synthesize
 192   user/group records for the `root` and `nobody` users if they are missing from
 193   `/etc/passwd`.
 194
 195 * `$SYSTEMD_NSS_DYNAMIC_BYPASS=1` — if set, `nss-systemd` won't return
 196   user/group records for dynamically registered service users (i.e. users
 197   registered through `DynamicUser=1`).
 198
 199 * `$SYSTEMD_NSS_BYPASS_BUS=1` — if set, `nss-systemd` won't use D-Bus to do
 200   dynamic user lookups. This is primarily useful to make `nss-systemd` work
 201   safely from within `dbus-daemon`.
 202
 203 systemd-timedated:
 204
 205 * `$SYSTEMD_TIMEDATED_NTP_SERVICES=…` — colon-separated list of unit names of
 206   NTP client services. If set, `timedatectl set-ntp on` enables and starts the
 207   first existing unit listed in the environment variable, and
 208   `timedatectl set-ntp off` disables and stops all listed units.
 209
 210 systemd-sulogin-shell:
 211
 212 * `$SYSTEMD_SULOGIN_FORCE=1` — This skips asking for the root password if the
 213   root password is not available (such as when the root account is locked).
 214   See `sulogin(8)` for more details.
 215
 216 bootctl and other tools that access the EFI System Partition (ESP):
 217
 218 * `$SYSTEMD_RELAX_ESP_CHECKS=1` — if set, the ESP validation checks are
 219   relaxed. Specifically, validation checks that ensure the specified ESP path
 220   is a FAT file system are turned off, as are checks that the path is located
 221   on a GPT partition with the correct type UUID.
 222
 223 * `$SYSTEMD_ESP_PATH=…` — override the path to the EFI System Partition. This
 224   may be used to override ESP path auto detection, and redirect any accesses to
 225   the ESP to the specified directory. Not that unlike with bootctl's --path=
 226   switch only very superficial validation of the specified path is done when
 227   this environment variable is used.
 228
 229 systemd itself:
 230
 231 * `$SYSTEMD_ACTIVATION_UNIT` — set for all NSS and PAM module invocations that
 232   are done by the service manager on behalf of a specific unit, in child
 233   processes that are later (after execve()) going to become unit
 234   processes. Contains the full unit name (e.g. "foobar.service"). NSS and PAM
 235   modules can use this information to determine in which context and on whose
 236   behalf they are being called, which may be useful to avoid deadlocks, for
 237   example to bypass IPC calls to the very service that is about to be
 238   started. Note that NSS and PAM modules should be careful to only rely on this
 239   data when invoked privileged, or possibly only when getppid() returns 1, as
 240   setting environment variables is of course possible in any even unprivileged
 241   contexts.
 242
 243 * `$SYSTEMD_ACTIVATION_SCOPE` — closely related to `$SYSTEMD_ACTIVATION_UNIT`,
 244   it is either set to `system` or `user` depending on whether the NSS/PAM
 245   module is called by systemd in `--system` or `--user` mode.
 246
 247 systemd-remount-fs:
 248
 249 * `$SYSTEMD_REMOUNT_ROOT_RW=1` — if set and no entry for the root directory
 250   exists in /etc/fstab (this file always takes precedence), then the root
 251   directory is remounted writable. This is primarily used by
 252   systemd-gpt-auto-generator to ensure the root partition is mounted writable
 253   in accordance to the GPT partition flags.
 254
 255 systemd-firstboot and localectl:
 256
 257 * `SYSTEMD_LIST_NON_UTF8_LOCALES=1` – if set non-UTF-8 locales are listed among
 258   the installed ones. By default non-UTF-8 locales are suppressed from the
 259   selection, since we are living in the 21st century.