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 * Libudev does no longer provide these symbols:
32 udev_monitor_from_socket()
33 udev_queue_get_failed_list_entry()
34 udev_get_{dev,sys,run}_path()
35 The verions number was bumped and symbol versioning introduced.
37 * systemd-logingctl and systemd-journalctl have been renamed
38 to logingctl and journalctl to match systemctl.
40 * The config files: /etc/systemd/systemd-logind.conf and
41 /etc/systemd/systemd-journald.conf have been renamed to
42 logind.conf and journald.conf. Package updates should rename
43 the files to the new names on upgrade.
46 * This is mostly a bugfix release
48 * Support optional initialization of the machine ID from the
49 KVM or container configured UUID.
51 * Support immediate reboots with "systemctl reboot -ff"
53 * Show /etc/os-release data in systemd-analyze output
55 * Many bugfixes for the journal, including endianess fixes and
56 ensuring that disk space enforcement works
58 * sd-login.h is C++ comptaible again
60 * Extend the /etc/os-release format on request of the Debian
63 * We now refuse non-UTF8 strings used in various configuration
64 and unit files. This is done to ensure we don't pass invalid
65 data over D-Bus or expose it elsewhere.
67 * Register Mimo USB Screens as suitable for automatic seat
70 * Read SELinux client context from journal clients in a race
73 * Reorder configuration file lookup order. /etc now always
74 overrides /run in order to allow the administrator to always
75 and unconditionally override vendor supplied or
76 automatically generated data.
78 * The various user visible bits of the journal now have man
79 pages. We still lack man pages for the journal API calls
82 * We now ship all man pages in HTML format again in the
85 Contributions from: Dave Reisner, Dirk Eibach, Frederic
86 Crozat, Harald Hoyer, Kay Sievers, Lennart Poettering, Marti
87 Raudsepp, Michal Schmidt, Shawn Landden, Tero Roponen, Thierry
91 * This is mostly a bugfix release
93 * systems lacking /etc/os-release are no longer supported.
95 * Various functionality updates to libsystemd-login.so
97 * Track class of PAM logins to distuingish greeters from
100 Contributions from: Kay Sievers, Lennart Poettering, Michael
104 * This is an important bugfix release for v41.
106 * Building man pages is now optional which should be useful
107 for those building systemd from git but unwilling to install
110 * Watchdog support for supervising services is now usable. In
111 a future release support for hardware watchdogs
112 (i.e. /dev/watchdog) will be added building on this.
114 * Service start rate limiting is now configurable and can be
115 turned off per service. When a start rate limit is hit a
116 reboot can automatically be triggered.
118 * New CanReboot(), CanPowerOff() bus calls in systemd-logind.
120 Contributions from: Benjamin Franzke, Bill Nottingham,
121 Frederic Crozat, Lennart Poettering, Michael Olbrich, Michal
122 Schmidt, Michał Górny, Piotr Drąg
125 * The systemd binary is installed /usr/lib/systemd/systemd now;
126 An existing /sbin/init symlink needs to be adapted with the
129 * The code that loads kernel modules has been ported to invoke
130 libkmod directly, instead of modprobe. This means we do not
131 support systems with module-init-tools anymore.
133 * Watchdog support is now already useful, but still not
136 * A new kernel command line option systemd.setenv= is
137 understood to set system wide environment variables
140 * We now limit the set of capabilities of systemd-journald.
142 * We now set SIGPIPE to ignore by default, since it only is
143 useful in shell pipelines, and has little use in general
144 code. This can be disabled with IgnoreSIPIPE=no in unit
147 Contributions from: Benjamin Franzke, Kay Sievers, Lennart
148 Poettering, Michael Olbrich, Michal Schmidt, Tom Gundersen,
152 * This is mostly a bugfix release
154 * We now expose the reason why a service failed in the
155 "Result" D-Bus property.
157 * Rudimentary service watchdog support (will be completed over
158 the next few releases.)
160 * When systemd forks off in order execute some service we will
161 now immediately changes its argv[0] to reflect which process
162 it will execute. This is useful to minimize the time window
163 with a generic argv[0], which makes bootcharts more useful
165 Contributions from: Alvaro Soliverez, Chris Paulson-Ellis, Kay
166 Sievers, Lennart Poettering, Michael Olbrich, Michal Schmidt,
167 Mike Kazantsev, Ray Strode
170 * This is mostly a test release, but incorporates many
173 * New systemd-cgtop tool to show control groups by their
176 * Linking against libacl for ACLs is optional again. If
177 disabled, support tracking device access for active logins
178 goes becomes unavailable, and so does access to the user
179 journals by the respective users.
181 * If a group "adm" exists, journal files are automatically
182 owned by them, thus allow members of this group full access
183 to the system journal as well as all user journals.
185 * The journal now stores the SELinux context of the logging
186 client for all entries.
188 * Add C++ inclusion guards to all public headers
190 * New output mode "cat" in the journal to print only text
191 messages, without any meta data like date or time.
193 * Include tiny X server wrapper as a temporary stop-gap to
194 teach XOrg udev display enumeration. This is used by display
195 managers such as gdm, and will go away as soon as XOrg
196 learned native udev hotplugging for display devices.
198 * Add new systemd-cat tool for executing arbitrary programs
199 with STDERR/STDOUT connected to the journal. Can also act as
200 BSD logger replacement, and does so by default.
202 * Optionally store all locally generated coredumps in the
203 journal along with meta data.
205 * systemd-tmpfiles learnt four new commands: n, L, c, b, for
206 writing short strings to files (for usage for /sys), and for
207 creating symlinks, character and block device nodes.
209 * New unit file option ControlGroupPersistent= to make cgroups
210 persistent, following the mechanisms outlined in
211 http://www.freedesktop.org/wiki/Software/systemd/PaxControlGroups
213 * Support multiple local RTCs in a sane way
215 * No longer monopolize IO when replaying readahead data on
216 rotating disks, since we might starve non-file-system IO to
217 death, since fanotify() will not see accesses done by blkid,
220 * Don't show kernel threads in systemd-cgls anymore, unless
221 requested with new -k switch.
223 Contributions from: Dan Horák, Kay Sievers, Lennart
224 Poettering, Michal Schmidt
227 * This is mostly a test release, but incorporates many
230 * The git repository moved to:
231 git://anongit.freedesktop.org/systemd/systemd
232 ssh://git.freedesktop.org/git/systemd/systemd
234 * First release with the journal
235 http://0pointer.de/blog/projects/the-journal.html
237 * The journal replaces both systemd-kmsg-syslogd and
238 systemd-stdout-bridge.
240 * New sd_pid_get_unit() API call in libsystemd-logind
242 * Many systemadm clean-ups
244 * Introduce remote-fs-pre.target which is ordered before all
245 remote mounts and may be used to start services before all
248 * Added Mageia support
250 * Add bash completion for systemd-loginctl
252 * Actively monitor PID file creation for daemons which exit in
253 the parent process before having finished writing the PID
254 file in the daemon process. Daemons which do this need to be
255 fixed (i.e. PID file creation must have finished before the
256 parent exits), but we now react a bit more gracefully to them.
258 * Add colourful boot output, mimicking the well-known output
259 of existing distributions.
261 * New option PassCredentials= for socket units, for
262 compatibility with a recent kernel ABI breakage.
264 * /etc/rc.local is now hooked in via a generator binary, and
265 thus will no longer act as synchronization point during
268 * systemctl list-unit-files now supports --root=.
270 * systemd-tmpfiles now understands two new commands: z, Z for
271 relabelling files according to the SELinux database. This is
272 useful to apply SELinux labels to specific files in /sys,
275 * Output of SysV services is now forwarded to both the console
276 and the journal by default, not only just the console.
278 * New man pages for all APIs from libsystemd-login.
280 * The build tree got reorganized and a the build system is a
281 lot more modular allowing embedded setups to specifically
282 select the components of systemd they are interested in.
284 * Support for Linux systems lacking the kernel VT subsystem is
287 * configure's --with-rootdir= got renamed to
288 --with-rootprefix= to follow the naming used by udev and
291 * Unless specified otherwise we'll now install to /usr instead
292 of /usr/local by default.
294 * Processes with '@' in argv[0][0] are now excluded from the
295 final shut-down killing spree, following the logic explained
297 http://www.freedesktop.org/wiki/Software/systemd/RootStorageDaemons
299 * All processes remaining in a service cgroup when we enter
300 the START or START_PRE states are now killed with
301 SIGKILL. That means it is no longer possible to spawn
302 background processes from ExecStart= lines (which was never
303 supported anyway, and bad style).
305 * New PropagateReloadTo=/PropagateReloadFrom= options to bind
306 reloading of units together.
308 Contributions from: Bill Nottingham, Daniel Walsh, Dave
309 Reisner, Dexter Morgan, Gregs Gregs, Jonathan Nieder, Kay
310 Sievers, Lennart Poettering, Michael Biebl, Michal Schmidt,
311 Michał Górny, Ran Benita, Thomas Jarosch, Tim Waugh, Tollef
312 Fog Heen, Tom Gundersen, Zbigniew Jędrzejewski-Szmek