1 systemd System and Service Manager
4 * This is mostly a bugfix release
6 * systems lacking /etc/os-release are no longer supported.
8 * Various functionality updates to libsystemd-login.so
10 * Track class of PAM logins to distuingish greeters from
13 Contributions from: Kay Sievers, Lennart Poettering, Michael
17 * This is an important bugfix release for v41.
19 * Building man pages is now optional which should be useful
20 for those building systemd from git but unwilling to install
23 * Watchdog support for supervising services is now usable. In
24 a future release support for hardware watchdogs
25 (i.e. /dev/watchdog) will be added building on this.
27 * Service start rate limiting is now configurable and can be
28 turned off per service. When a start rate limit is hit a
29 reboot can automatically be triggered.
31 * New CanReboot(), CanPowerOff() bus calls in systemd-logind.
33 Contributions from: Benjamin Franzke, Bill Nottingham,
34 Frederic Crozat, Lennart Poettering, Michael Olbrich, Michal
35 Schmidt, Michał Górny, Piotr Drąg
38 * The systemd binary is installed /usr/lib/systemd/systemd now;
39 An existing /sbin/init symlink needs to be adapted with the
42 * The code that loads kernel modules has been ported to invoke
43 libkmod directly, instead of modprobe. This means we do not
44 support systems with module-init-tools anymore.
46 * Watchdog support is now already useful, but still not
49 * A new kernel command line option systemd.setenv= is
50 understood to set system wide environment variables
53 * We now limit the set of capabilities of systemd-journald.
55 * We now set SIGPIPE to ignore by default, since it only is
56 useful in shell pipelines, and has little use in general
57 code. This can be disabled with IgnoreSIPIPE=no in unit
60 Contributions from: Benjamin Franzke, Kay Sievers, Lennart
61 Poettering, Michael Olbrich, Michal Schmidt, Tom Gundersen,
65 * This is mostly a bugfix release
67 * We now expose the reason why a service failed in the
68 "Result" D-Bus property.
70 * Rudimentary service watchdog support (will be completed over
71 the next few releases.)
73 * When systemd forks off in order execute some service we will
74 now immediately changes its argv[0] to reflect which process
75 it will execute. This is useful to minimize the time window
76 with a generic argv[0], which makes bootcharts more useful
78 Contributions from: Alvaro Soliverez, Chris Paulson-Ellis, Kay
79 Sievers, Lennart Poettering, Michael Olbrich, Michal Schmidt,
80 Mike Kazantsev, Ray Strode
83 * This is mostly a test release, but incorporates many
86 * New systemd-cgtop tool to show control groups by their
89 * Linking against libacl for ACLs is optional again. If
90 disabled, support tracking device access for active logins
91 goes becomes unavailable, and so does access to the user
92 journals by the respective users.
94 * If a group "adm" exists, journal files are automatically
95 owned by them, thus allow members of this group full access
96 to the system journal as well as all user journals.
98 * The journal now stores the SELinux context of the logging
99 client for all entries.
101 * Add C++ inclusion guards to all public headers
103 * New output mode "cat" in the journal to print only text
104 messages, without any meta data like date or time.
106 * Include tiny X server wrapper as a temporary stop-gap to
107 teach XOrg udev display enumeration. This is used by display
108 managers such as gdm, and will go away as soon as XOrg
109 learned native udev hotplugging for display devices.
111 * Add new systemd-cat tool for executing arbitrary programs
112 with STDERR/STDOUT connected to the journal. Can also act as
113 BSD logger replacement, and does so by default.
115 * Optionally store all locally generated coredumps in the
116 journal along with meta data.
118 * systemd-tmpfiles learnt four new commands: n, L, c, b, for
119 writing short strings to files (for usage for /sys), and for
120 creating symlinks, character and block device nodes.
122 * New unit file option ControlGroupPersistent= to make cgroups
123 persistent, following the mechanisms outlined in
124 http://www.freedesktop.org/wiki/Software/systemd/PaxControlGroups
126 * Support multiple local RTCs in a sane way
128 * No longer monopolize IO when replaying readahead data on
129 rotating disks, since we might starve non-file-system IO to
130 death, since fanotify() will not see accesses done by blkid,
133 * Don't show kernel threads in systemd-cgls anymore, unless
134 requested with new -k switch.
136 Contributions from: Dan Horák, Kay Sievers, Lennart
137 Poettering, Michal Schmidt
140 * This is mostly a test release, but incorporates many
143 * The git repository moved to:
144 git://anongit.freedesktop.org/systemd/systemd
145 ssh://git.freedesktop.org/git/systemd/systemd
147 * First release with the journal
148 http://0pointer.de/blog/projects/the-journal.html
150 * The journal replaces both systemd-kmsg-syslogd and
151 systemd-stdout-bridge.
153 * New sd_pid_get_unit() API call in libsystemd-logind
155 * Many systemadm clean-ups
157 * Introduce remote-fs-pre.target which is ordered before all
158 remote mounts and may be used to start services before all
161 * Added Mageia support
163 * Add bash completion for systemd-loginctl
165 * Actively monitor PID file creation for daemons which exit in
166 the parent process before having finished writing the PID
167 file in the daemon process. Daemons which do this need to be
168 fixed (i.e. PID file creation must have finished before the
169 parent exits), but we now react a bit more gracefully to them.
171 * Add colourful boot output, mimicking the well-known output
172 of existing distributions.
174 * New option PassCredentials= for socket units, for
175 compatibility with a recent kernel ABI breakage.
177 * /etc/rc.local is now hooked in via a generator binary, and
178 thus will no longer act as synchronization point during
181 * systemctl list-unit-files now supports --root=.
183 * systemd-tmpfiles now understands two new commands: z, Z for
184 relabelling files according to the SELinux database. This is
185 useful to apply SELinux labels to specific files in /sys,
188 * Output of SysV services is now forwarded to both the console
189 and the journal by default, not only just the console.
191 * New man pages for all APIs from libsystemd-login.
193 * The build tree got reorganized and a the build system is a
194 lot more modular allowing embedded setups to specifically
195 select the components of systemd they are interested in.
197 * Support for Linux systems lacking the kernel VT subsystem is
200 * configure's --with-rootdir= got renamed to
201 --with-rootprefix= to follow the naming used by udev and
204 * Unless specified otherwise we'll now install to /usr instead
205 of /usr/local by default.
207 * Processes with '@' in argv[0][0] are now excluded from the
208 final shut-down killing spree, following the logic explained
210 http://www.freedesktop.org/wiki/Software/systemd/RootStorageDaemons
212 * All processes remaining in a service cgroup when we enter
213 the START or START_PRE states are now killed with
214 SIGKILL. That means it is no longer possible to spawn
215 background processes from ExecStart= lines (which was never
216 supported anyway, and bad style).
218 * New PropagateReloadTo=/PropagateReloadFrom= options to bind
219 reloading of units together.
221 Contributions from: Bill Nottingham, Daniel Walsh, Dave
222 Reisner, Dexter Morgan, Gregs Gregs, Jonathan Nieder, Kay
223 Sievers, Lennart Poettering, Michael Biebl, Michal Schmidt,
224 Michał Górny, Ran Benita, Thomas Jarosch, Tim Waugh, Tollef
225 Fog Heen, Tom Gundersen, Zbigniew Jędrzejewski-Szmek