README

   1 systemd System and Service Manager
   2
   3 DETAILS:
   4         http://0pointer.de/blog/projects/systemd.html
   5
   6 WEB SITE:
   7         https://www.freedesktop.org/wiki/Software/systemd
   8
   9 GIT:
  10         git@github.com:systemd/systemd.git
  11         https://github.com/systemd/systemd.git
  12
  13 GITWEB:
  14         https://github.com/systemd/systemd
  15
  16 MAILING LIST:
  17         https://lists.freedesktop.org/mailman/listinfo/systemd-devel
  18
  19 IRC:
  20         #systemd on irc.freenode.org
  21
  22 BUG REPORTS:
  23         https://github.com/systemd/systemd/issues
  24
  25 AUTHOR:
  26         Lennart Poettering
  27         Kay Sievers
  28         ...and many others
  29
  30 LICENSE:
  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+
  36
  37 REQUIREMENTS:
  38         Linux kernel >= 3.13
  39         Linux kernel >= 4.2 for unified cgroup hierarchy support
  40
  41         Kernel Config Options:
  42           CONFIG_DEVTMPFS
  43           CONFIG_CGROUPS (it is OK to disable all controllers)
  44           CONFIG_INOTIFY_USER
  45           CONFIG_SIGNALFD
  46           CONFIG_TIMERFD
  47           CONFIG_EPOLL
  48           CONFIG_NET
  49           CONFIG_SYSFS
  50           CONFIG_PROC_FS
  51           CONFIG_FHANDLE (libudev, mount and bind mount handling)
  52
  53         Kernel crypto/hash API
  54           CONFIG_CRYPTO_USER_API_HASH
  55           CONFIG_CRYPTO_HMAC
  56           CONFIG_CRYPTO_SHA256
  57
  58         udev will fail to work with the legacy sysfs layout:
  59           CONFIG_SYSFS_DEPRECATED=n
  60
  61         Legacy hotplug slows down the system and confuses udev:
  62           CONFIG_UEVENT_HELPER_PATH=""
  63
  64         Userspace firmware loading is not supported and should
  65         be disabled in the kernel:
  66           CONFIG_FW_LOADER_USER_HELPER=n
  67
  68         Some udev rules and virtualization detection relies on it:
  69           CONFIG_DMIID
  70
  71         Support for some SCSI devices serial number retrieval, to
  72         create additional symlinks in /dev/disk/ and /dev/tape:
  73           CONFIG_BLK_DEV_BSG
  74
  75         Required for PrivateNetwork= and PrivateDevices= in service units:
  76           CONFIG_NET_NS
  77           CONFIG_DEVPTS_MULTIPLE_INSTANCES
  78         Note that systemd-localed.service and other systemd units use
  79         PrivateNetwork and PrivateDevices so this is effectively required.
  80
  81         Required for PrivateUsers= in service units:
  82           CONFIG_USER_NS
  83
  84         Optional but strongly recommended:
  85           CONFIG_IPV6
  86           CONFIG_AUTOFS4_FS
  87           CONFIG_TMPFS_XATTR
  88           CONFIG_{TMPFS,EXT4,XFS,BTRFS_FS,...}_POSIX_ACL
  89           CONFIG_SECCOMP
  90           CONFIG_SECCOMP_FILTER (required for seccomp support)
  91           CONFIG_CHECKPOINT_RESTORE (for the kcmp() syscall)
  92
  93         Required for CPUShares= in resource control unit settings
  94           CONFIG_CGROUP_SCHED
  95           CONFIG_FAIR_GROUP_SCHED
  96
  97         Required for CPUQuota= in resource control unit settings
  98           CONFIG_CFS_BANDWIDTH
  99
 100         For UEFI systems:
 101           CONFIG_EFIVAR_FS
 102           CONFIG_EFI_PARTITION
 103
 104         We recommend to turn off Real-Time group scheduling in the
 105         kernel when using systemd. RT group scheduling effectively
 106         makes RT scheduling unavailable for most userspace, since it
 107         requires explicit assignment of RT budgets to each unit whose
 108         processes making use of RT. As there's no sensible way to
 109         assign these budgets automatically this cannot really be
 110         fixed, and it's best to disable group scheduling hence.
 111            CONFIG_RT_GROUP_SCHED=n
 112
 113         Note that kernel auditing is broken when used with systemd's
 114         container code. When using systemd in conjunction with
 115         containers, please make sure to either turn off auditing at
 116         runtime using the kernel command line option "audit=0", or
 117         turn it off at kernel compile time using:
 118           CONFIG_AUDIT=n
 119         If systemd is compiled with libseccomp support on
 120         architectures which do not use socketcall() and where seccomp
 121         is supported (this effectively means x86-64 and ARM, but
 122         excludes 32-bit x86!), then nspawn will now install a
 123         work-around seccomp filter that makes containers boot even
 124         with audit being enabled. This works correctly only on kernels
 125         3.14 and newer though. TL;DR: turn audit off, still.
 126
 127         glibc >= 2.16
 128         libcap
 129         libmount >= 2.27.1 (from util-linux)
 130                 (util-linux *must* be built with --enable-libmount-force-mountinfo)
 131         libseccomp >= 2.3.1 (optional)
 132         libblkid >= 2.24 (from util-linux) (optional)
 133         libkmod >= 15 (optional)
 134         PAM >= 1.1.2 (optional)
 135         libcryptsetup (optional)
 136         libaudit (optional)
 137         libacl (optional)
 138         libselinux (optional)
 139         liblzma (optional)
 140         liblz4 >= 119 (optional)
 141         libgcrypt (optional)
 142         libqrencode (optional)
 143         libmicrohttpd (optional)
 144         libpython (optional)
 145         libidn (optional)
 146         elfutils >= 158 (optional)
 147         make, gcc, and similar tools
 148
 149         During runtime, you need the following additional
 150         dependencies:
 151
 152         util-linux >= v2.27.1 required
 153         dbus >= 1.4.0 (strictly speaking optional, but recommended)
 154                 NOTE: If using dbus < 1.9.18, you should override the default
 155                 policy directory (--with-dbuspolicydir=/etc/dbus-1/system.d).
 156         dracut (optional)
 157         PolicyKit (optional)
 158
 159         Two build systems are supported: meson + ninja-build and autools + make.
 160
 161         The following tools are needed with both systems:
 162
 163         pkg-config
 164         gperf >= 3.1
 165         docbook-xsl (optional, required for documentation)
 166         xsltproc    (optional, required for documentation)
 167         python-lxml (optional, required to build the indices)
 168
 169         When building with meson, python and ninja-build are required.
 170
 171         To build in directory build/:
 172           meson build/ && ninja -C build
 173
 174         Any configuration options can be specfied as -Darg=value... arguments
 175         to meson. After the build directory is initially configured, meson will
 176         refuse to run again, and options must be changed with:
 177           mesonconf -Darg=value...
 178         mesonconf without any arguments will print out available options and
 179         their current values.
 180
 181         Useful commands:
 182           ninja -v some/target
 183           ninja test
 184           sudo ninja install
 185           DESTDIR=... ninja install
 186
 187         When building with autotools, the following tools are needed:
 188
 189         automake
 190         autoconf
 191         libtool
 192         intltool
 193         python (optional)
 194
 195         The build system is initialized with ./autogen.sh and the usual
 196           ./configure && make
 197         should be used.
 198
 199         A tar ball can be created with:
 200           git archive --format=tar --prefix=systemd-222/ v222 | xz > systemd-222.tar.xz
 201
 202         When systemd-hostnamed is used, it is strongly recommended to
 203         install nss-myhostname to ensure that, in a world of
 204         dynamically changing hostnames, the hostname stays resolvable
 205         under all circumstances. In fact, systemd-hostnamed will warn
 206         if nss-myhostname is not installed.
 207
 208         Additional packages are necessary to run some tests:
 209         - busybox            (used by test/TEST-13-NSPAWN-SMOKE)
 210         - nc                 (used by test/TEST-12-ISSUE-3171)
 211         - python3-pyparsing
 212         - python3-evdev      (used by hwdb parsing tests)
 213         - strace             (used by test/test-functions)
 214         - capsh              (optional, used by test-execute)
 215
 216 USERS AND GROUPS:
 217         Default udev rules use the following standard system group
 218         names, which need to be resolvable by getgrnam() at any time,
 219         even in the very early boot stages, where no other databases
 220         and network are available:
 221
 222         audio, cdrom, dialout, disk, input, kmem, lp, tape, tty, video
 223
 224         During runtime, the journal daemon requires the
 225         "systemd-journal" system group to exist. New journal files will
 226         be readable by this group (but not writable), which may be used
 227         to grant specific users read access. In addition, system
 228         groups "wheel" and "adm" will be given read-only access to
 229         journal files using systemd-tmpfiles.service.
 230
 231         The journal gateway daemon requires the
 232         "systemd-journal-gateway" system user and group to
 233         exist. During execution this network facing service will drop
 234         privileges and assume this uid/gid for security reasons.
 235
 236         Similarly, the NTP daemon requires the "systemd-timesync" system
 237         user and group to exist.
 238
 239         Similarly, the network management daemon requires the
 240         "systemd-network" system user and group to exist.
 241
 242         Similarly, the name resolution daemon requires the
 243         "systemd-resolve" system user and group to exist.
 244
 245         Similarly, the coredump support requires the
 246         "systemd-coredump" system user and group to exist.
 247
 248 NSS:
 249         systemd ships with four glibc NSS modules:
 250
 251         nss-myhostname resolves the local hostname to locally
 252         configured IP addresses, as well as "localhost" to
 253         127.0.0.1/::1.
 254
 255         nss-resolve enables DNS resolution via the systemd-resolved
 256         DNS/LLMNR caching stub resolver "systemd-resolved".
 257
 258         nss-mymachines enables resolution of all local containers registered
 259         with machined to their respective IP addresses. It also maps UID/GIDs
 260         ranges used by containers to useful names.
 261
 262         nss-systemd enables resolution of all dynamically allocated service
 263         users. (See the DynamicUser= setting in unit files.)
 264
 265         To make use of these NSS modules, please add them to the "hosts:",
 266         "passwd:" and "group:" lines in /etc/nsswitch.conf. The "resolve"
 267         module should replace the glibc "dns" module in this file (and don't
 268         worry, it chain-loads the "dns" module if it can't talk to resolved).
 269
 270         The four modules should be used in the following order:
 271
 272                 passwd: compat mymachines systemd
 273                 group: compat mymachines systemd
 274                 hosts: files mymachines resolve myhostname
 275
 276 SYSV INIT.D SCRIPTS:
 277         When calling "systemctl enable/disable/is-enabled" on a unit which is a
 278         SysV init.d script, it calls /usr/lib/systemd/systemd-sysv-install;
 279         this needs to translate the action into the distribution specific
 280         mechanism such as chkconfig or update-rc.d. Packagers need to provide
 281         this script if you need this functionality (you don't if you disabled
 282         SysV init support).
 283
 284         Please see src/systemctl/systemd-sysv-install.SKELETON for how this
 285         needs to look like, and provide an implementation at the marked places.
 286
 287 WARNINGS:
 288         systemd will warn you during boot if /usr is on a different
 289         file system than /. While in systemd itself very little will
 290         break if /usr is on a separate partition, many of its
 291         dependencies very likely will break sooner or later in one
 292         form or another. For example, udev rules tend to refer to
 293         binaries in /usr, binaries that link to libraries in /usr or
 294         binaries that refer to data files in /usr. Since these
 295         breakages are not always directly visible, systemd will warn
 296         about this, since this kind of file system setup is not really
 297         supported anymore by the basic set of Linux OS components.
 298
 299         systemd requires that the /run mount point exists. systemd also
 300         requires that /var/run is a symlink to /run.
 301
 302         For more information on this issue consult
 303         https://www.freedesktop.org/wiki/Software/systemd/separate-usr-is-broken
 304
 305         To run systemd under valgrind, compile with VALGRIND defined
 306         (e.g. ./configure CPPFLAGS='... -DVALGRIND=1'). Otherwise,
 307         false positives will be triggered by code which violates
 308         some rules but is actually safe.
 309
 310 ENGINEERING AND CONSULTING SERVICES:
 311         Kinvolk (https://kinvolk.io) offers professional engineering
 312         and consulting services for systemd. Please contact Chris Kühl
 313         <chris@kinvolk.io> for more information.