README

   1 systemd System and Service Manager
   2
   3 DETAILS:
   4         http://0pointer.de/blog/projects/systemd.html
   5
   6 WEB SITE:
   7         http://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         http://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 >= 3.12
  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         udev will fail to work with the legacy sysfs layout:
  54           CONFIG_SYSFS_DEPRECATED=n
  55
  56         Legacy hotplug slows down the system and confuses udev:
  57           CONFIG_UEVENT_HELPER_PATH=""
  58
  59         Userspace firmware loading is not supported and should
  60         be disabled in the kernel:
  61           CONFIG_FW_LOADER_USER_HELPER=n
  62
  63         Some udev rules and virtualization detection relies on it:
  64           CONFIG_DMIID
  65
  66         Support for some SCSI devices serial number retrieval, to
  67         create additional symlinks in /dev/disk/ and /dev/tape:
  68           CONFIG_BLK_DEV_BSG
  69
  70         Required for PrivateNetwork and PrivateDevices in service units:
  71           CONFIG_NET_NS
  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.
  75
  76         Optional but strongly recommended:
  77           CONFIG_IPV6
  78           CONFIG_AUTOFS4_FS
  79           CONFIG_TMPFS_XATTR
  80           CONFIG_{TMPFS,EXT4,XFS,BTRFS_FS,...}_POSIX_ACL
  81           CONFIG_SECCOMP
  82           CONFIG_CHECKPOINT_RESTORE (for the kcmp() syscall)
  83
  84         Required for CPUShares= in resource control unit settings
  85           CONFIG_CGROUP_SCHED
  86           CONFIG_FAIR_GROUP_SCHED
  87
  88         Required for CPUQuota= in resource control unit settings
  89           CONFIG_CFS_BANDWIDTH
  90
  91         For UEFI systems:
  92           CONFIG_EFIVAR_FS
  93           CONFIG_EFI_PARTITION
  94
  95         We recommend to turn off Real-Time group scheduling in the
  96         kernel when using systemd. RT group scheduling effectively
  97         makes RT scheduling unavailable for most userspace, since it
  98         requires explicit assignment of RT budgets to each unit whose
  99         processes making use of RT. As there's no sensible way to
 100         assign these budgets automatically this cannot really be
 101         fixed, and it's best to disable group scheduling hence.
 102            CONFIG_RT_GROUP_SCHED=n
 103
 104         Note that kernel auditing is broken when used with systemd's
 105         container code. When using systemd in conjunction with
 106         containers, please make sure to either turn off auditing at
 107         runtime using the kernel command line option "audit=0", or
 108         turn it off at kernel compile time using:
 109           CONFIG_AUDIT=n
 110         If systemd is compiled with libseccomp support on
 111         architectures which do not use socketcall() and where seccomp
 112         is supported (this effectively means x86-64 and ARM, but
 113         excludes 32-bit x86!), then nspawn will now install a
 114         work-around seccomp filter that makes containers boot even
 115         with audit being enabled. This works correctly only on kernels
 116         3.14 and newer though. TL;DR: turn audit off, still.
 117
 118         glibc >= 2.16
 119         libcap
 120         libmount >= 2.27.1 (from util-linux)
 121                 (util-linux *must* be built with --enable-libmount-force-mountinfo)
 122         libseccomp >= 1.0.0 (optional)
 123         libblkid >= 2.24 (from util-linux) (optional)
 124         libkmod >= 15 (optional)
 125         PAM >= 1.1.2 (optional)
 126         libcryptsetup (optional)
 127         libaudit (optional)
 128         libacl (optional)
 129         libselinux (optional)
 130         liblzma (optional)
 131         liblz4 >= 119 (optional)
 132         libgcrypt (optional)
 133         libqrencode (optional)
 134         libmicrohttpd (optional)
 135         libpython (optional)
 136         libidn (optional)
 137         elfutils >= 158 (optional)
 138         make, gcc, and similar tools
 139
 140         During runtime, you need the following additional
 141         dependencies:
 142
 143         util-linux >= v2.27.1 required
 144         dbus >= 1.4.0 (strictly speaking optional, but recommended)
 145         dracut (optional)
 146         PolicyKit (optional)
 147
 148         When building from git, the following tools are needed:
 149
 150         pkg-config
 151         docbook-xsl
 152         xsltproc
 153         automake
 154         autoconf
 155         libtool
 156         intltool
 157         gperf
 158         python (optional)
 159         python-lxml (optional, but required to build the indices)
 160
 161         The build system is initialized with ./autogen.sh. A tar ball
 162         can be created with:
 163           git archive --format=tar --prefix=systemd-222/ v222 | xz > systemd-222.tar.xz
 164
 165         When systemd-hostnamed is used, it is strongly recommended to
 166         install nss-myhostname to ensure that, in a world of
 167         dynamically changing hostnames, the hostname stays resolvable
 168         under all circumstances. In fact, systemd-hostnamed will warn
 169         if nss-myhostname is not installed.
 170
 171 USERS AND GROUPS:
 172         Default udev rules use the following standard system group
 173         names, which need to be resolvable by getgrnam() at any time,
 174         even in the very early boot stages, where no other databases
 175         and network are available:
 176
 177         audio, cdrom, dialout, disk, input, kmem, lp, tape, tty, video
 178
 179         During runtime, the journal daemon requires the
 180         "systemd-journal" system group to exist. New journal files will
 181         be readable by this group (but not writable), which may be used
 182         to grant specific users read access. In addition, system
 183         groups "wheel" and "adm" will be given read-only access to
 184         journal files using systemd-tmpfiles.service.
 185
 186         The journal gateway daemon requires the
 187         "systemd-journal-gateway" system user and group to
 188         exist. During execution this network facing service will drop
 189         privileges and assume this uid/gid for security reasons.
 190
 191         Similarly, the NTP daemon requires the "systemd-timesync" system
 192         user and group to exist.
 193
 194         Similarly, the network management daemon requires the
 195         "systemd-network" system user and group to exist.
 196
 197         Similarly, the name resolution daemon requires the
 198         "systemd-resolve" system user and group to exist.
 199
 200         Similarly, the coredump support requires the
 201         "systemd-coredump" system user and group to exist.
 202
 203 NSS:
 204         systemd ships with four glibc NSS modules:
 205
 206         nss-myhostname resolves the local hostname to locally
 207         configured IP addresses, as well as "localhost" to
 208         127.0.0.1/::1.
 209
 210         nss-resolve enables DNS resolution via the systemd-resolved
 211         DNS/LLMNR caching stub resolver "systemd-resolved".
 212
 213         nss-mymachines enables resolution of all local containers registered
 214         with machined to their respective IP addresses. It also maps UID/GIDs
 215         ranges used by containers to useful names.
 216
 217         nss-systemd enables resolution of all dynamically allocated service
 218         users. (See the DynamicUser= setting in unit files.)
 219
 220         To make use of these NSS modules, please add them to the "hosts:",
 221         "passwd:" and "group:" lines in /etc/nsswitch.conf. The "resolve"
 222         module should replace the glibc "dns" module in this file (and don't
 223         worry, it chain-loads the "dns" module if it can't talk to resolved).
 224
 225         The four modules should be used in the following order:
 226
 227                 passwd: compat mymachines systemd
 228                 group: compat mymachines systemd
 229                 hosts: files mymachines resolve myhostname
 230
 231 SYSV INIT.D SCRIPTS:
 232         When calling "systemctl enable/disable/is-enabled" on a unit which is a
 233         SysV init.d script, it calls /usr/lib/systemd/systemd-sysv-install;
 234         this needs to translate the action into the distribution specific
 235         mechanism such as chkconfig or update-rc.d. Packagers need to provide
 236         this script if you need this functionality (you don't if you disabled
 237         SysV init support).
 238
 239         Please see src/systemctl/systemd-sysv-install.SKELETON for how this
 240         needs to look like, and provide an implementation at the marked places.
 241
 242 WARNINGS:
 243         systemd will warn you during boot if /usr is on a different
 244         file system than /. While in systemd itself very little will
 245         break if /usr is on a separate partition, many of its
 246         dependencies very likely will break sooner or later in one
 247         form or another. For example, udev rules tend to refer to
 248         binaries in /usr, binaries that link to libraries in /usr or
 249         binaries that refer to data files in /usr. Since these
 250         breakages are not always directly visible, systemd will warn
 251         about this, since this kind of file system setup is not really
 252         supported anymore by the basic set of Linux OS components.
 253
 254         systemd requires that the /run mount point exists. systemd also
 255         requires that /var/run is a symlink to /run.
 256
 257         For more information on this issue consult
 258         http://freedesktop.org/wiki/Software/systemd/separate-usr-is-broken
 259
 260         To run systemd under valgrind, compile with VALGRIND defined
 261         (e.g. ./configure CPPFLAGS='... -DVALGRIND=1'). Otherwise,
 262         false positives will be triggered by code which violates
 263         some rules but is actually safe.
 264
 265         Currently, systemd-timesyncd defaults to use the Google NTP
 266         servers if not specified otherwise at configure time. You
 267         really should not ship an OS or device with this default
 268         setting. See DISTRO_PORTING for details.
 269
 270 ENGINEERING AND CONSULTING SERVICES:
 271         Kinvolk (https://kinvolk.io) offers professional engineering
 272         and consulting services for systemd. Please contact Chris Kühl
 273         <chris@kinvolk.io> for more information.