docs/ENVIRONMENT.md

   1 ---
   2 title: Known Environment Variables
   3 ---
   4
   5 # Known Environment Variables
   6
   7 A number of systemd components take additional runtime parameters via
   8 environment variables. Many of these environment variables are not supported at
   9 the same level as command line switches and other interfaces are: we don't
  10 document them in the man pages and we make no stability guarantees for
  11 them. While they generally are unlikely to be dropped any time soon again, we
  12 do not want to guarantee that they stay around for good either.
  13
  14 Below is an (incomprehensive) list of the environment variables understood by
  15 the various tools. Note that this list only covers environment variables not
  16 documented in the proper man pages.
  17
  18 All tools:
  19
  20 * `$SYSTEMD_OFFLINE=[0|1]` — if set to `1`, then `systemctl` will
  21   refrain from talking to PID 1; this has the same effect as the historical
  22   detection of `chroot()`.  Setting this variable to `0` instead has a similar
  23   effect as `SYSTEMD_IGNORE_CHROOT=1`; i.e. tools will try to
  24   communicate with PID 1 even if a `chroot()` environment is detected.
  25   You almost certainly want to set this to `1` if you maintain a package build system
  26   or similar and are trying to use a modern container system and not plain
  27   `chroot()`.
  28
  29 * `$SYSTEMD_IGNORE_CHROOT=1` — if set, don't check whether being invoked in a
  30   `chroot()` environment. This is particularly relevant for systemctl, as it
  31   will not alter its behaviour for `chroot()` environments if set.  Normally it
  32   refrains from talking to PID 1 in such a case; turning most operations such
  33   as `start` into no-ops.  If that's what's explicitly desired, you might
  34   consider setting `SYSTEMD_OFFLINE=1`.
  35
  36 * `$SD_EVENT_PROFILE_DELAYS=1` — if set, the sd-event event loop implementation
  37   will print latency information at runtime.
  38
  39 * `$SYSTEMD_PROC_CMDLINE` — if set, may contain a string that is used as kernel
  40   command line instead of the actual one readable from /proc/cmdline. This is
  41   useful for debugging, in order to test generators and other code against
  42   specific kernel command lines.
  43
  44 * `$SYSTEMD_IN_INITRD` — takes a boolean. If set, overrides initrd detection.
  45   This is useful for debugging and testing initrd-only programs in the main
  46   system.
  47
  48 * `$SYSTEMD_BUS_TIMEOUT=SECS` — specifies the maximum time to wait for method call
  49   completion. If no time unit is specified, assumes seconds. The usual other units
  50   are understood, too (us, ms, s, min, h, d, w, month, y). If it is not set or set
  51   to 0, then the built-in default is used.
  52
  53 * `$SYSTEMD_MEMPOOL=0` — if set, the internal memory caching logic employed by
  54   hash tables is turned off, and libc malloc() is used for all allocations.
  55
  56 * `$SYSTEMD_EMOJI=0` — if set, tools such as "systemd-analyze security" will
  57   not output graphical smiley emojis, but ASCII alternatives instead. Note that
  58   this only controls use of Unicode emoji glyphs, and has no effect on other
  59   Unicode glyphs.
  60
  61 systemctl:
  62
  63 * `$SYSTEMCTL_FORCE_BUS=1` — if set, do not connect to PID1's private D-Bus
  64   listener, and instead always connect through the dbus-daemon D-bus broker.
  65
  66 * `$SYSTEMCTL_INSTALL_CLIENT_SIDE=1` — if set, enable or disable unit files on
  67   the client side, instead of asking PID 1 to do this.
  68
  69 * `$SYSTEMCTL_SKIP_SYSV=1` — if set, do not call out to SysV compatibility hooks.
  70
  71 systemd-nspawn:
  72
  73 * `$UNIFIED_CGROUP_HIERARCHY=1` — if set, force nspawn into unified cgroup
  74   hierarchy mode.
  75
  76 * `$SYSTEMD_NSPAWN_API_VFS_WRITABLE=1` — if set, make /sys and /proc/sys and
  77   friends writable in the container. If set to "network", leave only
  78   /proc/sys/net writable.
  79
  80 * `$SYSTEMD_NSPAWN_CONTAINER_SERVICE=…` — override the "service" name nspawn
  81   uses to register with machined. If unset defaults to "nspawn", but with this
  82   variable may be set to any other value.
  83
  84 * `$SYSTEMD_NSPAWN_USE_CGNS=0` — if set, do not use cgroup namespacing, even if
  85   it is available.
  86
  87 * `$SYSTEMD_NSPAWN_LOCK=0` — if set, do not lock container images when running.
  88
  89 * `$SYSTEMD_NSPAWN_TMPFS_TMP=0` — if set, do not overmount /tmp in the
  90   container with a tmpfs, but leave the directory from the image in place.
  91
  92 systemd-logind:
  93
  94 * `$SYSTEMD_BYPASS_HIBERNATION_MEMORY_CHECK=1` — if set, report that
  95   hibernation is available even if the swap devices do not provide enough room
  96   for it.
  97
  98 systemd-udevd:
  99
 100 * `$NET_NAMING_SCHEME=` – if set, takes a network naming scheme (i.e. one of
 101   "v238", "v239", "v240"…, or the special value "latest") as parameter. If
 102   specified udev's net_id builtin will follow the specified naming scheme when
 103   determining stable network interface names. This may be used to revert to
 104   naming schemes of older udev versions, in order to provide more stable naming
 105   across updates. This environment variable takes precedence over the kernel
 106   command line option `net.naming-scheme=`, except if the value is prefixed
 107   with `:` in which case the kernel command line option takes precedence, if it
 108   is specified as well.
 109
 110 * `$SYSTEMD_REBOOT_TO_FIRMWARE_SETUP` — if set overrides systemd-logind's
 111   built-in EFI logic of requesting a reboot into the firmware. Takes a
 112   boolean. If set to false the functionality is turned off entirely. If set to
 113   true instead of requesting a reboot into the firmware setup UI through EFI a
 114   file `/run/systemd/reboot-to-firmware-setup` is created whenever this is
 115   requested. This file may be checked for by services run during system
 116   shutdown in order to request the appropriate operation from the firmware in
 117   an alternative fashion.
 118
 119 * `$SYSTEMD_REBOOT_TO_BOOT_LOADER_MENU` — similar to the above, allows
 120   overriding of systemd-logind's built-in EFI logic of requesting a reboot into
 121   the boot loader menu. Takes a boolean. If set to false the functionality is
 122   turned off entirely. If set to true instead of requesting a reboot into the
 123   boot loader menu through EFI a file `/run/systemd/reboot-to-boot-loader-menu`
 124   is created whenever this is requested. The file contains the requested boot
 125   loader menu timeout in µs, formatted in ASCII decimals, or zero in case no
 126   time-out is requested. This file may be checked for by services run during
 127   system shutdown in order to request the appropriate operation from the boot
 128   loader in an alternative fashion.
 129
 130 * `$SYSTEMD_REBOOT_TO_BOOT_LOADER_ENTRY` — similar to the above, allows
 131   overriding of systemd-logind's built-in EFI logic of requesting a reboot into
 132   a specific boot loader entry. Takes a boolean. If set to false the
 133   functionality is turned off entirely. If set to true instead of requesting a
 134   reboot into a specific boot loader entry through EFI a file
 135   `/run/systemd/reboot-to-boot-loader-entry` is created whenever this is
 136   requested. The file contains the requested boot loader entry identifier. This
 137   file may be checked for by services run during system shutdown in order to
 138   request the appropriate operation from the boot loader in an alternative
 139   fashion. Note that by default only boot loader entries which follow the [Boot
 140   Loader Specification](https://systemd.io/BOOT_LOADER_SPECIFICATION) and are
 141   placed in the ESP or the Extended Boot Loader partition may be selected this
 142   way. However, if a directory `/run/boot-loader-entries/` exists, the entries
 143   are loaded from there instead. The directory should contain the usual
 144   directory hierarchy mandated by the Boot Loader Specification, i.e. the entry
 145   drop-ins should be placed in
 146   `/run/boot-loader-entries/loader/entries/*.conf`, and the files referenced by
 147   the drop-ins (including the kernels and initrds) somewhere else below
 148   `/run/boot-loader-entries/`. Note that all these files may be (and are
 149   supposed to be) symlinks. systemd-logind will load these files on-demand,
 150   these files can hence be updated (ideally atomically) whenever the boot
 151   loader configuration changes. A foreign boot loader installer script should
 152   hence synthesize drop-in snippets and symlinks for all boot entries at boot
 153   or whenever they change if it wants to integrate with systemd-logind's APIs.
 154
 155 installed systemd tests:
 156
 157 * `$SYSTEMD_TEST_DATA` — override the location of test data. This is useful if
 158   a test executable is moved to an arbitrary location.
 159
 160 nss-systemd:
 161
 162 * `$SYSTEMD_NSS_BYPASS_SYNTHETIC=1` — if set, `nss-systemd` won't synthesize
 163   user/group records for the `root` and `nobody` users if they are missing from
 164   `/etc/passwd`.
 165
 166 * `$SYSTEMD_NSS_DYNAMIC_BYPASS=1` — if set, `nss-systemd` won't return
 167   user/group records for dynamically registered service users (i.e. users
 168   registered through `DynamicUser=1`).
 169
 170 * `$SYSTEMD_NSS_BYPASS_BUS=1` — if set, `nss-systemd` won't use D-Bus to do
 171   dynamic user lookups. This is primarily useful to make `nss-systemd` work
 172   safely from within `dbus-daemon`.
 173
 174 systemd-timedated:
 175
 176 * `$SYSTEMD_TIMEDATED_NTP_SERVICES=…` — colon-separated list of unit names of
 177   NTP client services. If set, `timedatectl set-ntp on` enables and starts the
 178   first existing unit listed in the environment variable, and
 179   `timedatectl set-ntp off` disables and stops all listed units.
 180
 181 systemd-sulogin-shell:
 182
 183 * `$SYSTEMD_SULOGIN_FORCE=1` — This skips asking for the root password if the
 184   root password is not available (such as when the root account is locked).
 185   See `sulogin(8)` for more details.
 186
 187 bootctl and other tools that access the EFI System Partition (ESP):
 188
 189 * `$SYSTEMD_RELAX_ESP_CHECKS=1` — if set, the ESP validation checks are
 190   relaxed. Specifically, validation checks that ensure the specified ESP path
 191   is a FAT file system are turned off, as are checks that the path is located
 192   on a GPT partition with the correct type UUID.
 193
 194 * `$SYSTEMD_ESP_PATH=…` — override the path to the EFI System Partition. This
 195   may be used to override ESP path auto detection, and redirect any accesses to
 196   the ESP to the specified directory. Not that unlike with bootctl's --path=
 197   switch only very superficial validation of the specified path is done when
 198   this environment variable is used.
 199
 200 systemd itself:
 201
 202 * `$SYSTEMD_ACTIVATION_UNIT` — set for all NSS and PAM module invocations that
 203   are done by the service manager on behalf of a specific unit, in child
 204   processes that are later (after execve()) going to become unit
 205   processes. Contains the full unit name (e.g. "foobar.service"). NSS and PAM
 206   modules can use this information to determine in which context and on whose
 207   behalf they are being called, which may be useful to avoid deadlocks, for
 208   example to bypass IPC calls to the very service that is about to be
 209   started. Note that NSS and PAM modules should be careful to only rely on this
 210   data when invoked privileged, or possibly only when getppid() returns 1, as
 211   setting environment variables is of course possible in any even unprivileged
 212   contexts.
 213
 214 * `$SYSTEMD_ACTIVATION_SCOPE` — closely related to `$SYSTEMD_ACTIVATION_UNIT`,
 215   it is either set to `system` or `user` depending on whether the NSS/PAM
 216   module is called by systemd in `--system` or `--user` mode.
 217
 218 systemd-remount-fs:
 219
 220 * `$SYSTEMD_REMOUNT_ROOT_RW=1` — if set and and no entry for the root directory
 221   exists in /etc/fstab (this file always takes precedence), then the root
 222   directory is remounted writable. This is primarily used by
 223   systemd-gpt-auto-generator to ensure the root partition is mounted writable
 224   in accordance to the GPT partition flags.