1 systemd System and Service Manager
7 git@github.com:systemd/systemd.git
8 https://github.com/systemd/systemd
11 https://lists.freedesktop.org/mailman/listinfo/systemd-devel
14 #systemd on irc.libera.chat
17 https://github.com/systemd/systemd/issues
20 https://0pointer.de/blog/projects/systemd.html
21 https://www.freedesktop.org/wiki/Software/systemd
29 LGPL-2.1-or-later for all code, exceptions noted in LICENSES/README.md
33 ≥ 4.3 for ambient capabilities
34 ≥ 4.5 for pids controller in cgroup v2
35 ≥ 4.6 for cgroup namespaces
36 ≥ 4.9 for RENAME_NOREPLACE support in vfat
37 ≥ 4.10 for cgroup-bpf egress and ingress hooks
38 ≥ 4.15 for cgroup-bpf device hook and cpu controller in cgroup v2
39 ≥ 4.17 for cgroup-bpf socket address hooks
40 ≥ 4.20 for PSI (used by systemd-oomd)
41 ≥ 5.3 for bounded loops in BPF program
42 ≥ 5.4 for signed Verity images
43 ≥ 5.7 for BPF links and the BPF LSM hook
45 ⛔ Kernel versions below 3.15 ("minimum baseline") are not supported at
46 all, and are missing required functionality (e.g. CLOCK_BOOTTIME
47 support for timerfd_create()).
49 ⚠️ Kernel versions below 4.15 ("recommended baseline") have significant
50 gaps in functionality and are not recommended for use with this version
51 of systemd (e.g. lack sufficiently comprehensive and working cgroupv2
52 support). Taint flag 'old-kernel' will be set. systemd will most likely
53 still function, but upstream support and testing are limited.
55 Kernel Config Options:
57 CONFIG_CGROUPS (it is OK to disable all controllers)
62 CONFIG_UNIX (it requires CONFIG_NET, but every other flag in it is not necessary)
65 CONFIG_FHANDLE (libudev, mount and bind mount handling)
67 udev will fail to work with the legacy sysfs layout:
68 CONFIG_SYSFS_DEPRECATED=n
70 Legacy hotplug slows down the system and confuses udev:
71 CONFIG_UEVENT_HELPER_PATH=""
73 Userspace firmware loading is not supported and should be disabled in
75 CONFIG_FW_LOADER_USER_HELPER=n
77 Some udev rules and virtualization detection relies on it:
80 Support for some SCSI devices serial number retrieval, to create
81 additional symlinks in /dev/disk/ and /dev/tape:
84 Required for PrivateNetwork= in service units:
86 Note that systemd-localed.service and other systemd units use
87 PrivateNetwork so this is effectively required.
89 Required for PrivateUsers= in service units:
92 Optional but strongly recommended:
96 CONFIG_{TMPFS,EXT4_FS,XFS,BTRFS_FS,...}_POSIX_ACL
98 CONFIG_SECCOMP_FILTER (required for seccomp support)
99 CONFIG_KCMP (for the kcmp() syscall, used to be under
100 CONFIG_CHECKPOINT_RESTORE before ~5.12)
102 CONFIG_NET_SCH_FQ_CODEL
104 Required for CPUShares= in resource control unit settings:
106 CONFIG_FAIR_GROUP_SCHED
108 Required for CPUQuota= in resource control unit settings:
111 Required for IPAddressDeny=, IPAddressAllow=, IPIngressFilterPath=,
112 IPEgressFilterPath= in resource control unit settings unit settings:
119 Required for SocketBind{Allow|Deny}=, RestrictNetworkInterfaces= in
120 resource control unit settings:
131 Required for signed Verity images support:
132 CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG
133 Required to verify signed Verity images using keys enrolled in the MoK
134 (Machine-Owner Key) keyring:
135 CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG_SECONDARY_KEYRING
136 CONFIG_IMA_ARCH_POLICY
137 CONFIG_INTEGRITY_MACHINE_KEYRING
139 Required for reading credentials from SMBIOS:
143 Required for RestrictFileSystems= in service units:
147 CONFIG_DEBUG_INFO_BTF
148 CONFIG_LSM="...,bpf" or kernel booted with lsm="...,bpf".
150 We recommend to turn off Real-Time group scheduling in the kernel when
151 using systemd. RT group scheduling effectively makes RT scheduling
152 unavailable for most userspace, since it requires explicit assignment of
153 RT budgets to each unit whose processes making use of RT. As there's no
154 sensible way to assign these budgets automatically this cannot really be
155 fixed, and it's best to disable group scheduling hence:
156 CONFIG_RT_GROUP_SCHED=n
158 It's a good idea to disable the implicit creation of networking bonding
159 devices by the kernel networking bonding module, so that the
160 automatically created "bond0" interface doesn't conflict with any such
161 device created by systemd-networkd (or other tools). Ideally there would
162 be a kernel compile-time option for this, but there currently isn't. The
163 next best thing is to make this change through a modprobe.d drop-in.
164 This is shipped by default, see modprobe.d/systemd.conf.
166 Required for systemd-nspawn:
167 CONFIG_DEVPTS_MULTIPLE_INSTANCES or Linux kernel >= 4.7
169 Required for systemd-oomd:
172 Note that kernel auditing is broken when used with systemd's container
173 code. When using systemd in conjunction with containers, please make
174 sure to either turn off auditing at runtime using the kernel command
175 line option "audit=0", or turn it off at kernel compile time using:
178 If systemd is compiled with libseccomp support on architectures which do
179 not use socketcall() and where seccomp is supported (this effectively
180 means x86-64 and ARM, but excludes 32-bit x86!), then nspawn will now
181 install a work-around seccomp filter that makes containers boot even
182 with audit being enabled. This works correctly only on kernels 3.14 and
183 newer though. TL;DR: turn audit off, still.
187 libmount >= 2.30 (from util-linux)
188 (util-linux *must* be built without --enable-libmount-support-mtab)
189 libseccomp >= 2.3.1 (optional)
190 libblkid >= 2.24 (from util-linux) (optional)
191 libkmod >= 15 (optional)
192 PAM >= 1.1.2 (optional)
193 libcryptsetup (optional), >= 2.3.0 required for signed Verity images support
196 libbpf >= 0.1.0 (optional)
197 libfdisk >= 2.32 (from util-linux) (optional)
198 libselinux (optional)
200 liblz4 >= 1.3.0 / 130 (optional)
201 libzstd >= 1.4.0 (optional)
203 libqrencode (optional)
204 libmicrohttpd (optional)
205 libidn2 or libidn (optional)
206 gnutls >= 3.1.4 (optional, >= 3.6.0 is required to support DNS-over-TLS with gnutls)
207 openssl >= 1.1.0 (optional, required to support DNS-over-TLS with openssl)
208 elfutils >= 158 (optional)
210 tzdata >= 2014f (optional)
213 docbook-xsl (optional, required for documentation)
214 xsltproc (optional, required for documentation)
215 python >= 3.7 (required by meson too, >= 3.9 is required for ukify)
217 python-pefile (optional, required for ukify)
218 python-lxml (optional, required to build the indices)
219 pyelftools (optional, required for systemd-boot)
223 awk, sed, grep, and similar tools
224 clang >= 10.0, llvm >= 10.0 (optional, required to build BPF programs
225 from source code in C)
227 During runtime, you need the following additional
230 util-linux >= v2.27.1 required (including but not limited to: mount,
231 umount, swapon, swapoff, sulogin,
233 dbus >= 1.4.0 (strictly speaking optional, but recommended)
234 NOTE: If using dbus < 1.9.18, you should override the default
235 policy directory (--with-dbuspolicydir=/etc/dbus-1/system.d).
238 To build in directory build/:
239 meson setup build/ && ninja -C build/
241 Any configuration options can be specified as -Darg=value... arguments
242 to meson. After the build directory is initially configured, meson will
243 refuse to run again, and options must be changed with:
244 meson configure -Darg=value build/
245 meson configure without any arguments will print out available options and
246 their current values.
249 ninja -C build -v some/target
251 sudo meson install -C build/ --no-rebuild
252 DESTDIR=... meson install -C build/
254 A tarball can be created with:
255 v=250 && git archive --prefix=systemd-$v/ v$v | zstd >systemd-$v.tar.zstd
257 When systemd-hostnamed is used, it is strongly recommended to install
258 nss-myhostname to ensure that, in a world of dynamically changing
259 hostnames, the hostname stays resolvable under all circumstances. In
260 fact, systemd-hostnamed will warn if nss-myhostname is not installed.
262 nss-systemd must be enabled on systemd systems, as that's required for
263 DynamicUser= to work. Note that we ship services out-of-the-box that
264 make use of DynamicUser= now, hence enabling nss-systemd is not
267 Note that the build prefix for systemd must be /usr/. (Moreover, packages
268 systemd relies on — such as D-Bus — really should use the same prefix,
269 otherwise you are on your own.) Split-usr and unmerged-usr systems are no
270 longer supported, and moving everything under /usr/ is required. Systems
271 with a separate /usr/ partition must mount it before transitioning into it
272 (i.e.: from the initrd). For more information see:
273 https://www.freedesktop.org/wiki/Software/systemd/separate-usr-is-broken
274 https://www.freedesktop.org/wiki/Software/systemd/TheCaseForTheUsrMerge
276 Additional packages are necessary to run some tests:
277 - nc (used by test/TEST-12-ISSUE-3171)
278 - python (test-udev which is installed is in python)
280 - python-evdev (used by hwdb parsing tests)
281 - strace (used by test/test-functions)
282 - capsh (optional, used by test-execute)
284 POLICY FOR SUPPORT OF DISTRIBUTIONS AND ARCHITECTURES:
285 systemd main branch and latest major or stable releases are generally
286 expected to compile on current versions of popular distributions (at
287 least all non-EOL versions of Fedora, Debian unstable/testing/stable,
288 latest Ubuntu LTS and non-LTS releases, openSUSE Tumbleweed/Leap,
289 CentOS Stream 8 and 9, up-to-date Arch, etc.) We will generally
290 attempt to support also other non-EOL versions of various distros.
291 Features which would break compilation on slightly older distributions
292 will only be introduced if there are significant reasons for this
293 (i.e. supporting them interferes with development or requires too many
294 resources to support). In some cases backports of specific libraries or
295 tools might be required.
297 The policy is similar for architecture support. systemd is regularly
298 tested on popular architectures (currently amd64, i386, arm64, ppc64el,
299 and s390x), but should compile and work also on other architectures, for
300 which support has been added. systemd will emit warnings when
301 architecture-specific constants are not defined.
303 STATIC COMPILATION AND "STANDALONE" BINARIES:
304 systemd provides a public shared libraries libsystemd.so and
305 libudev.so. The latter is deprecated, and the sd-device APIs in
306 libsystemd should be used instead for new code. In addition, systemd is
307 built with a private shared library, libsystemd-shared-<suffix>.so,
308 that also includes the libsystemd code, and by default most systemd
309 binaries are linked to it. Using shared libraries saves disk space and
310 memory at runtime, because only one copy of the code is needed.
312 It is possible to build static versions of systemd public shared
313 libraries (via the configuration options '-Dstatic-libsystemd' and
314 '-Dstatic-libudev'). This allows the libsystemd and libudev code to be
315 linked statically into programs. Note that mixing & matching different
316 versions of libsystemd and systemd is generally not recommended, since
317 various of its APIs wrap internal state and protocols of systemd
318 (e.g. logind and udev databases), which are not considered
319 stable. Hence, using static libraries is not recommended since it
320 generally means that version of the static libsystemd linked into
321 applications and the host systemd are not in sync, and will thus create
322 compatibility problems.
324 In addition, it is possible to disable the use of
325 libsystemd-shared-<suffix>.so for various components (via the
326 configuration options '-Dlink-*-shared'). In this mode, the libsystemd
327 and libsystemd-shared code is linked statically into selected
328 binaries. This option is intended for systems where some of the
329 components are intended to be delivered independently of the main
330 systemd package. Finally, some binaries can be compiled in a second
331 version (via the configuration option '-Dstandalone-binaries'). The
332 version suffixed with ".standalone" has the libsystemd and
333 libsystemd-shared code linked statically. Those binaries are intended
334 as replacements to be used in limited installations where the full
335 systemd suite is not installed. Yet another option is to rebuild
336 systemd with a different '-Dshared-lib-tag' setting, allowing different
337 systemd binaries to be linked to instances of the private shared
338 library that can be installed in parallel.
340 Again: Using the default shared linking is recommended, building static
341 or "standalone" versions is not. Mixing versions of systemd components
342 that would normally be built and used together (in particular various
343 daemons and the manager) is not recommended: we do not test such
344 combinations upstream and cannot provide support. Distributors making
345 use of those options are responsible if things do not work as expected.
348 Default udev rules use the following standard system group names, which
349 need to be resolvable by getgrnam() at any time, even in the very early
350 boot stages, where no other databases and network are available:
352 audio, cdrom, dialout, disk, input, kmem, kvm, lp, render, tape, tty, video
354 During runtime, the journal daemon requires the "systemd-journal" system
355 group to exist. New journal files will be readable by this group (but
356 not writable), which may be used to grant specific users read access. In
357 addition, system groups "wheel" and "adm" will be given read-only access
358 to journal files using systemd-tmpfiles-setup.service.
360 The journal remote daemon requires the "systemd-journal-remote" system
361 user and group to exist. During execution this network facing service
362 will drop privileges and assume this uid/gid for security reasons.
364 Similarly, the network management daemon requires the "systemd-network"
365 system user and group to exist.
367 Similarly, the name resolution daemon requires the "systemd-resolve"
368 system user and group to exist.
370 Similarly, the coredump support requires the "systemd-coredump" system
371 user and group to exist.
374 systemd ships with four glibc NSS modules:
376 nss-myhostname resolves the local hostname to locally configured IP
377 addresses, as well as "localhost" to 127.0.0.1/::1.
379 nss-resolve enables DNS resolution via the systemd-resolved DNS/LLMNR
380 caching stub resolver "systemd-resolved".
382 nss-mymachines enables resolution of all local containers registered
383 with machined to their respective IP addresses.
385 nss-systemd enables resolution of users/group registered via the
386 User/Group Record Lookup API (https://systemd.io/USER_GROUP_API),
387 including all dynamically allocated service users. (See the
388 DynamicUser= setting in unit files.)
390 To make use of these NSS modules, please add them to the "hosts:",
391 "passwd:", "group:", "shadow:" and "gshadow:" lines in
394 The four modules should be used in the following order:
396 passwd: files systemd
397 group: files [SUCCESS=merge] systemd
398 shadow: files systemd
399 gshadow: files systemd
400 hosts: mymachines resolve [!UNAVAIL=return] files myhostname dns
403 When calling "systemctl enable/disable/is-enabled" on a unit which is a
404 SysV init.d script, it calls /usr/lib/systemd/systemd-sysv-install;
405 this needs to translate the action into the distribution specific
406 mechanism such as chkconfig or update-rc.d. Packagers need to provide
407 this script if you need this functionality (you don't if you disabled
410 Please see src/systemctl/systemd-sysv-install.SKELETON for how this
411 needs to look like, and provide an implementation at the marked places.
413 WARNINGS and TAINT FLAGS:
414 systemd requires that the /run mount point exists. systemd also
415 requires that /var/run is a symlink to /run. Taint flag 'var-run-bad'
416 will be set when this condition is detected.
418 Systemd will also warn when the cgroup support is unavailable in the
419 kernel (taint flag 'cgroups-missing'), the system is using the old
420 cgroup hierarchy (taint flag 'cgroupsv1'), the hardware clock is
421 running in non-UTC mode (taint flag 'local-hwclock'), the kernel
422 overflow UID or GID are not 65534 (taint flags 'overflowuid-not-65534'
423 and 'overflowgid-not-65534'), the UID or GID range assigned to the
424 running systemd instance covers less than 0…65534 (taint flags
425 'short-uid-range' and 'short-gid-range').
427 Taint conditions are logged during boot, but may also be checked at any
430 busctl get-property org.freedesktop.systemd1 /org/freedesktop/systemd1 org.freedesktop.systemd1.Manager Tainted
432 See org.freedesktop.systemd1(5) for more information.
435 To run systemd under valgrind, compile systemd with the valgrind
436 development headers available (i.e. valgrind-devel or equivalent).
437 Otherwise, false positives will be triggered by code which violates
438 some rules but is actually safe. Note that valgrind generates nice
439 output only on exit(), hence on shutdown we don't execve()
442 STABLE BRANCHES AND BACKPORTS:
443 Stable branches with backported patches are available in the
444 systemd-stable repo at https://github.com/systemd/systemd-stable.
446 Stable branches are started for certain releases of systemd and named
447 after them, e.g. v238-stable. Stable branches are managed by
448 distribution maintainers on an as needed basis. See
449 https://www.freedesktop.org/wiki/Software/systemd/Backports for some
450 more information and examples.