1 systemd System and Service Manager
4 * udev: all udev sources are merged into the systemd source tree now.
5 All future udev development will happen in the systemd tree. It
6 is still fully supported to use the udev daemon and tools without
7 systemd running, like in initramfs or other init systems. Building
8 udev though, will require the *build* of the systemd tree, but
9 udev can be properly *run* without systems.
11 * udev: /lib/udev/devices/ are not read anymore; systemd-tmpfiles
12 should be used to create dead device nodes as workarounds for broken
15 * udev: RUN+="socket:..." and udev_monitor_new_from_socket() is
16 no longer supported. udev_monitor_new_from_netlink() needs to be
17 used to subscribe to events.
19 * udev: when udevd is started by systemd, processes which are left
20 behind by forking them off of udev rules, are unconditionally cleaned
21 up and killed now after the event handling has finished. Services or
22 daemons must be started as systemd services. Services can be
23 pulled-in by udev to ge started, but they can no longer be directly
26 * udev: the daemon binary is called systemd-udevd now and installed
27 in /usr/lib/systemd/. Standalone builds or non-systemd systems need
28 to adapt to that, create symlink, or rename the binary after building
31 * systemd-logingctl and systemd-journalctl have been renamed
32 to logingctl and journalctl to match systemctl.
34 * The config files: /etc/systemd/systemd-logind.conf and
35 /etc/systemd/systemd-journald.conf have been renamed to
36 logind.conf and journald.conf. Package updates should rename
37 the files to the new names on upgrade.
40 * This is mostly a bugfix release
42 * Support optional initialization of the machine ID from the
43 KVM or container configured UUID.
45 * Support immediate reboots with "systemctl reboot -ff"
47 * Show /etc/os-release data in systemd-analyze output
49 * Many bugfixes for the journal, including endianess fixes and
50 ensuring that disk space enforcement works
52 * sd-login.h is C++ comptaible again
54 * Extend the /etc/os-release format on request of the Debian
57 * We now refuse non-UTF8 strings used in various configuration
58 and unit files. This is done to ensure we don't pass invalid
59 data over D-Bus or expose it elsewhere.
61 * Register Mimo USB Screens as suitable for automatic seat
64 * Read SELinux client context from journal clients in a race
67 * Reorder configuration file lookup order. /etc now always
68 overrides /run in order to allow the administrator to always
69 and unconditionally override vendor supplied or
70 automatically generated data.
72 * The various user visible bits of the journal now have man
73 pages. We still lack man pages for the journal API calls
76 * We now ship all man pages in HTML format again in the
79 Contributions from: Dave Reisner, Dirk Eibach, Frederic
80 Crozat, Harald Hoyer, Kay Sievers, Lennart Poettering, Marti
81 Raudsepp, Michal Schmidt, Shawn Landden, Tero Roponen, Thierry
85 * This is mostly a bugfix release
87 * systems lacking /etc/os-release are no longer supported.
89 * Various functionality updates to libsystemd-login.so
91 * Track class of PAM logins to distuingish greeters from
94 Contributions from: Kay Sievers, Lennart Poettering, Michael
98 * This is an important bugfix release for v41.
100 * Building man pages is now optional which should be useful
101 for those building systemd from git but unwilling to install
104 * Watchdog support for supervising services is now usable. In
105 a future release support for hardware watchdogs
106 (i.e. /dev/watchdog) will be added building on this.
108 * Service start rate limiting is now configurable and can be
109 turned off per service. When a start rate limit is hit a
110 reboot can automatically be triggered.
112 * New CanReboot(), CanPowerOff() bus calls in systemd-logind.
114 Contributions from: Benjamin Franzke, Bill Nottingham,
115 Frederic Crozat, Lennart Poettering, Michael Olbrich, Michal
116 Schmidt, Michał Górny, Piotr Drąg
119 * The systemd binary is installed /usr/lib/systemd/systemd now;
120 An existing /sbin/init symlink needs to be adapted with the
123 * The code that loads kernel modules has been ported to invoke
124 libkmod directly, instead of modprobe. This means we do not
125 support systems with module-init-tools anymore.
127 * Watchdog support is now already useful, but still not
130 * A new kernel command line option systemd.setenv= is
131 understood to set system wide environment variables
134 * We now limit the set of capabilities of systemd-journald.
136 * We now set SIGPIPE to ignore by default, since it only is
137 useful in shell pipelines, and has little use in general
138 code. This can be disabled with IgnoreSIPIPE=no in unit
141 Contributions from: Benjamin Franzke, Kay Sievers, Lennart
142 Poettering, Michael Olbrich, Michal Schmidt, Tom Gundersen,
146 * This is mostly a bugfix release
148 * We now expose the reason why a service failed in the
149 "Result" D-Bus property.
151 * Rudimentary service watchdog support (will be completed over
152 the next few releases.)
154 * When systemd forks off in order execute some service we will
155 now immediately changes its argv[0] to reflect which process
156 it will execute. This is useful to minimize the time window
157 with a generic argv[0], which makes bootcharts more useful
159 Contributions from: Alvaro Soliverez, Chris Paulson-Ellis, Kay
160 Sievers, Lennart Poettering, Michael Olbrich, Michal Schmidt,
161 Mike Kazantsev, Ray Strode
164 * This is mostly a test release, but incorporates many
167 * New systemd-cgtop tool to show control groups by their
170 * Linking against libacl for ACLs is optional again. If
171 disabled, support tracking device access for active logins
172 goes becomes unavailable, and so does access to the user
173 journals by the respective users.
175 * If a group "adm" exists, journal files are automatically
176 owned by them, thus allow members of this group full access
177 to the system journal as well as all user journals.
179 * The journal now stores the SELinux context of the logging
180 client for all entries.
182 * Add C++ inclusion guards to all public headers
184 * New output mode "cat" in the journal to print only text
185 messages, without any meta data like date or time.
187 * Include tiny X server wrapper as a temporary stop-gap to
188 teach XOrg udev display enumeration. This is used by display
189 managers such as gdm, and will go away as soon as XOrg
190 learned native udev hotplugging for display devices.
192 * Add new systemd-cat tool for executing arbitrary programs
193 with STDERR/STDOUT connected to the journal. Can also act as
194 BSD logger replacement, and does so by default.
196 * Optionally store all locally generated coredumps in the
197 journal along with meta data.
199 * systemd-tmpfiles learnt four new commands: n, L, c, b, for
200 writing short strings to files (for usage for /sys), and for
201 creating symlinks, character and block device nodes.
203 * New unit file option ControlGroupPersistent= to make cgroups
204 persistent, following the mechanisms outlined in
205 http://www.freedesktop.org/wiki/Software/systemd/PaxControlGroups
207 * Support multiple local RTCs in a sane way
209 * No longer monopolize IO when replaying readahead data on
210 rotating disks, since we might starve non-file-system IO to
211 death, since fanotify() will not see accesses done by blkid,
214 * Don't show kernel threads in systemd-cgls anymore, unless
215 requested with new -k switch.
217 Contributions from: Dan Horák, Kay Sievers, Lennart
218 Poettering, Michal Schmidt
221 * This is mostly a test release, but incorporates many
224 * The git repository moved to:
225 git://anongit.freedesktop.org/systemd/systemd
226 ssh://git.freedesktop.org/git/systemd/systemd
228 * First release with the journal
229 http://0pointer.de/blog/projects/the-journal.html
231 * The journal replaces both systemd-kmsg-syslogd and
232 systemd-stdout-bridge.
234 * New sd_pid_get_unit() API call in libsystemd-logind
236 * Many systemadm clean-ups
238 * Introduce remote-fs-pre.target which is ordered before all
239 remote mounts and may be used to start services before all
242 * Added Mageia support
244 * Add bash completion for systemd-loginctl
246 * Actively monitor PID file creation for daemons which exit in
247 the parent process before having finished writing the PID
248 file in the daemon process. Daemons which do this need to be
249 fixed (i.e. PID file creation must have finished before the
250 parent exits), but we now react a bit more gracefully to them.
252 * Add colourful boot output, mimicking the well-known output
253 of existing distributions.
255 * New option PassCredentials= for socket units, for
256 compatibility with a recent kernel ABI breakage.
258 * /etc/rc.local is now hooked in via a generator binary, and
259 thus will no longer act as synchronization point during
262 * systemctl list-unit-files now supports --root=.
264 * systemd-tmpfiles now understands two new commands: z, Z for
265 relabelling files according to the SELinux database. This is
266 useful to apply SELinux labels to specific files in /sys,
269 * Output of SysV services is now forwarded to both the console
270 and the journal by default, not only just the console.
272 * New man pages for all APIs from libsystemd-login.
274 * The build tree got reorganized and a the build system is a
275 lot more modular allowing embedded setups to specifically
276 select the components of systemd they are interested in.
278 * Support for Linux systems lacking the kernel VT subsystem is
281 * configure's --with-rootdir= got renamed to
282 --with-rootprefix= to follow the naming used by udev and
285 * Unless specified otherwise we'll now install to /usr instead
286 of /usr/local by default.
288 * Processes with '@' in argv[0][0] are now excluded from the
289 final shut-down killing spree, following the logic explained
291 http://www.freedesktop.org/wiki/Software/systemd/RootStorageDaemons
293 * All processes remaining in a service cgroup when we enter
294 the START or START_PRE states are now killed with
295 SIGKILL. That means it is no longer possible to spawn
296 background processes from ExecStart= lines (which was never
297 supported anyway, and bad style).
299 * New PropagateReloadTo=/PropagateReloadFrom= options to bind
300 reloading of units together.
302 Contributions from: Bill Nottingham, Daniel Walsh, Dave
303 Reisner, Dexter Morgan, Gregs Gregs, Jonathan Nieder, Kay
304 Sievers, Lennart Poettering, Michael Biebl, Michal Schmidt,
305 Michał Górny, Ran Benita, Thomas Jarosch, Tim Waugh, Tollef
306 Fog Heen, Tom Gundersen, Zbigniew Jędrzejewski-Szmek