1 systemd System and Service Manager
4 http://0pointer.de/blog/projects/systemd.html
7 http://www.freedesktop.org/wiki/Software/systemd
10 git@github.com:systemd/systemd.git
11 https://github.com/systemd/systemd.git
14 https://github.com/systemd/systemd
17 http://lists.freedesktop.org/mailman/listinfo/systemd-devel
20 #systemd on irc.freenode.org
23 https://github.com/systemd/systemd/issues
31 LGPLv2.1+ for all code
32 - except src/basic/MurmurHash2.c which is Public Domain
33 - except src/basic/siphash24.c which is CC0 Public Domain
34 - except src/journal/lookup3.c which is Public Domain
35 - except src/udev/* which is (currently still) GPLv2, GPLv2+
39 Linux kernel >= 4.2 for unified cgroup hierarchy support
41 Kernel Config Options:
43 CONFIG_CGROUPS (it is OK to disable all controllers)
51 CONFIG_FHANDLE (libudev, mount and bind mount handling)
53 udev will fail to work with the legacy sysfs layout:
54 CONFIG_SYSFS_DEPRECATED=n
56 Legacy hotplug slows down the system and confuses udev:
57 CONFIG_UEVENT_HELPER_PATH=""
59 Userspace firmware loading is not supported and should
60 be disabled in the kernel:
61 CONFIG_FW_LOADER_USER_HELPER=n
63 Some udev rules and virtualization detection relies on it:
66 Support for some SCSI devices serial number retrieval, to
67 create additional symlinks in /dev/disk/ and /dev/tape:
70 Required for PrivateNetwork= and PrivateDevices= in service units:
72 CONFIG_DEVPTS_MULTIPLE_INSTANCES
73 Note that systemd-localed.service and other systemd units use
74 PrivateNetwork and PrivateDevices so this is effectively required.
76 Required for PrivateUsers= in service units:
79 Optional but strongly recommended:
83 CONFIG_{TMPFS,EXT4,XFS,BTRFS_FS,...}_POSIX_ACL
85 CONFIG_SECCOMP_FILTER (required for seccomp support)
86 CONFIG_CHECKPOINT_RESTORE (for the kcmp() syscall)
88 Required for CPUShares= in resource control unit settings
90 CONFIG_FAIR_GROUP_SCHED
92 Required for CPUQuota= in resource control unit settings
99 We recommend to turn off Real-Time group scheduling in the
100 kernel when using systemd. RT group scheduling effectively
101 makes RT scheduling unavailable for most userspace, since it
102 requires explicit assignment of RT budgets to each unit whose
103 processes making use of RT. As there's no sensible way to
104 assign these budgets automatically this cannot really be
105 fixed, and it's best to disable group scheduling hence.
106 CONFIG_RT_GROUP_SCHED=n
108 Note that kernel auditing is broken when used with systemd's
109 container code. When using systemd in conjunction with
110 containers, please make sure to either turn off auditing at
111 runtime using the kernel command line option "audit=0", or
112 turn it off at kernel compile time using:
114 If systemd is compiled with libseccomp support on
115 architectures which do not use socketcall() and where seccomp
116 is supported (this effectively means x86-64 and ARM, but
117 excludes 32-bit x86!), then nspawn will now install a
118 work-around seccomp filter that makes containers boot even
119 with audit being enabled. This works correctly only on kernels
120 3.14 and newer though. TL;DR: turn audit off, still.
124 libmount >= 2.27.1 (from util-linux)
125 (util-linux *must* be built with --enable-libmount-force-mountinfo)
126 libseccomp >= 2.3.1 (optional)
127 libblkid >= 2.24 (from util-linux) (optional)
128 libkmod >= 15 (optional)
129 PAM >= 1.1.2 (optional)
130 libcryptsetup (optional)
133 libselinux (optional)
135 liblz4 >= 119 (optional)
137 libqrencode (optional)
138 libmicrohttpd (optional)
141 elfutils >= 158 (optional)
142 make, gcc, and similar tools
144 During runtime, you need the following additional
147 util-linux >= v2.27.1 required
148 dbus >= 1.4.0 (strictly speaking optional, but recommended)
149 NOTE: If using dbus < 1.9.18, you should override the default
150 policy directory (--with-dbuspolicydir=/etc/dbus-1/system.d).
154 When building from git, the following tools are needed:
165 python-lxml (optional, but required to build the indices)
167 The build system is initialized with ./autogen.sh. A tar ball
169 git archive --format=tar --prefix=systemd-222/ v222 | xz > systemd-222.tar.xz
171 When systemd-hostnamed is used, it is strongly recommended to
172 install nss-myhostname to ensure that, in a world of
173 dynamically changing hostnames, the hostname stays resolvable
174 under all circumstances. In fact, systemd-hostnamed will warn
175 if nss-myhostname is not installed.
177 Additional packages are necessary to run some tests:
178 - busybox (used by test/TEST-13-NSPAWN-SMOKE)
179 - nc (used by test/TEST-12-ISSUE-3171)
181 - python3-evdev (used by hwdb parsing tests)
182 - strace (used by test/test-functions)
183 - capsh (optional, used by test-execute)
186 Default udev rules use the following standard system group
187 names, which need to be resolvable by getgrnam() at any time,
188 even in the very early boot stages, where no other databases
189 and network are available:
191 audio, cdrom, dialout, disk, input, kmem, lp, tape, tty, video
193 During runtime, the journal daemon requires the
194 "systemd-journal" system group to exist. New journal files will
195 be readable by this group (but not writable), which may be used
196 to grant specific users read access. In addition, system
197 groups "wheel" and "adm" will be given read-only access to
198 journal files using systemd-tmpfiles.service.
200 The journal gateway daemon requires the
201 "systemd-journal-gateway" system user and group to
202 exist. During execution this network facing service will drop
203 privileges and assume this uid/gid for security reasons.
205 Similarly, the NTP daemon requires the "systemd-timesync" system
206 user and group to exist.
208 Similarly, the network management daemon requires the
209 "systemd-network" system user and group to exist.
211 Similarly, the name resolution daemon requires the
212 "systemd-resolve" system user and group to exist.
214 Similarly, the coredump support requires the
215 "systemd-coredump" system user and group to exist.
218 systemd ships with four glibc NSS modules:
220 nss-myhostname resolves the local hostname to locally
221 configured IP addresses, as well as "localhost" to
224 nss-resolve enables DNS resolution via the systemd-resolved
225 DNS/LLMNR caching stub resolver "systemd-resolved".
227 nss-mymachines enables resolution of all local containers registered
228 with machined to their respective IP addresses. It also maps UID/GIDs
229 ranges used by containers to useful names.
231 nss-systemd enables resolution of all dynamically allocated service
232 users. (See the DynamicUser= setting in unit files.)
234 To make use of these NSS modules, please add them to the "hosts:",
235 "passwd:" and "group:" lines in /etc/nsswitch.conf. The "resolve"
236 module should replace the glibc "dns" module in this file (and don't
237 worry, it chain-loads the "dns" module if it can't talk to resolved).
239 The four modules should be used in the following order:
241 passwd: compat mymachines systemd
242 group: compat mymachines systemd
243 hosts: files mymachines resolve myhostname
246 When calling "systemctl enable/disable/is-enabled" on a unit which is a
247 SysV init.d script, it calls /usr/lib/systemd/systemd-sysv-install;
248 this needs to translate the action into the distribution specific
249 mechanism such as chkconfig or update-rc.d. Packagers need to provide
250 this script if you need this functionality (you don't if you disabled
253 Please see src/systemctl/systemd-sysv-install.SKELETON for how this
254 needs to look like, and provide an implementation at the marked places.
257 systemd will warn you during boot if /usr is on a different
258 file system than /. While in systemd itself very little will
259 break if /usr is on a separate partition, many of its
260 dependencies very likely will break sooner or later in one
261 form or another. For example, udev rules tend to refer to
262 binaries in /usr, binaries that link to libraries in /usr or
263 binaries that refer to data files in /usr. Since these
264 breakages are not always directly visible, systemd will warn
265 about this, since this kind of file system setup is not really
266 supported anymore by the basic set of Linux OS components.
268 systemd requires that the /run mount point exists. systemd also
269 requires that /var/run is a symlink to /run.
271 For more information on this issue consult
272 http://freedesktop.org/wiki/Software/systemd/separate-usr-is-broken
274 To run systemd under valgrind, compile with VALGRIND defined
275 (e.g. ./configure CPPFLAGS='... -DVALGRIND=1'). Otherwise,
276 false positives will be triggered by code which violates
277 some rules but is actually safe.
279 ENGINEERING AND CONSULTING SERVICES:
280 Kinvolk (https://kinvolk.io) offers professional engineering
281 and consulting services for systemd. Please contact Chris Kühl
282 <chris@kinvolk.io> for more information.