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