1 systemd System and Service Manager
4 * The systemd binary is installed /usr/lib/systemd/systemd now;
5 An existing /sbin/init symlink needs to be adapted with the
8 * The code that loads kernel modules has been ported to invoke
9 libkmod directly, instead of modprobe. This means we do not
10 support systems with module-init-tools anymore.
12 * Watchdog support is now already useful, but still not
15 * A new kernel command line option systemd.setenv= is
16 understood to set system wide environment variables
19 * We now limit the set of capabilities of systemd-journald.
21 * We now set SIGPIPE to ignore by default, since it only is
22 useful in shell pipelines, and has little use in general
23 code. This can be disabled with IgnoreSIPIPE=no in unit
26 Contributions from: Benjamin Franzke, Kay Sievers, Lennart
27 Poettering, Michael Olbrich, Michal Schmidt, Tom Gundersen,
31 * This is mostly a bugfix release
33 * We now expose the reason why a service failed in the
34 "Result" D-Bus property.
36 * Rudimentary service watchdog support (will be completed over
37 the next few releases.)
39 * When systemd forks off in order execute some service we will
40 now immediately changes its argv[0] to reflect which process
41 it will execute. This is useful to minimize the time window
42 with a generic argv[0], which makes bootcharts more useful
44 Contributions from: Alvaro Soliverez, Chris Paulson-Ellis, Kay
45 Sievers, Lennart Poettering, Michael Olbrich, Michal Schmidt,
46 Mike Kazantsev, Ray Strode
49 * This is mostly a test release, but incorporates many
52 * New systemd-cgtop tool to show control groups by their
55 * Linking against libacl for ACLs is optional again. If
56 disabled, support tracking device access for active logins
57 goes becomes unavailable, and so does access to the user
58 journals by the respective users.
60 * If a group "adm" exists, journal files are automatically
61 owned by them, thus allow members of this group full access
62 to the system journal as well as all user journals.
64 * The journal now stores the SELinux context of the logging
65 client for all entries.
67 * Add C++ inclusion guards to all public headers
69 * New output mode "cat" in the journal to print only text
70 messages, without any meta data like date or time.
72 * Include tiny X server wrapper as a temporary stop-gap to
73 teach XOrg udev display enumeration. This is used by display
74 managers such as gdm, and will go away as soon as XOrg
75 learned native udev hotplugging for display devices.
77 * Add new systemd-cat tool for executing arbitrary programs
78 with STDERR/STDOUT connected to the journal. Can also act as
79 BSD logger replacement, and does so by default.
81 * Optionally store all locally generated coredumps in the
82 journal along with meta data.
84 * systemd-tmpfiles learnt four new commands: n, L, c, b, for
85 writing short strings to files (for usage for /sys), and for
86 creating symlinks, character and block device nodes.
88 * New unit file option ControlGroupPersistent= to make cgroups
89 persistent, following the mechanisms outlined in
90 http://www.freedesktop.org/wiki/Software/systemd/PaxControlGroups
92 * Support multiple local RTCs in a sane way
94 * No longer monopolize IO when replaying readahead data on
95 rotating disks, since we might starve non-file-system IO to
96 death, since fanotify() will not see accesses done by blkid,
99 * Don't show kernel threads in systemd-cgls anymore, unless
100 requested with new -k switch.
102 Contributions from: Dan Horák, Kay Sievers, Lennart
103 Poettering, Michal Schmidt
106 * This is mostly a test release, but incorporates many
109 * The git repository moved to:
110 git://anongit.freedesktop.org/systemd/systemd
111 ssh://git.freedesktop.org/git/systemd/systemd
113 * First release with the journal
114 http://0pointer.de/blog/projects/the-journal.html
116 * The journal replaces both systemd-kmsg-syslogd and
117 systemd-stdout-bridge.
119 * New sd_pid_get_unit() API call in libsystemd-logind
121 * Many systemadm clean-ups
123 * Introduce remote-fs-pre.target which is ordered before all
124 remote mounts and may be used to start services before all
127 * Added Mageia support
129 * Add bash completion for systemd-loginctl
131 * Actively monitor PID file creation for daemons which exit in
132 the parent process before having finished writing the PID
133 file in the daemon process. Daemons which do this need to be
134 fixed (i.e. PID file creation must have finished before the
135 parent exits), but we now react a bit more gracefully to them.
137 * Add colourful boot output, mimicking the well-known output
138 of existing distributions.
140 * New option PassCredentials= for socket units, for
141 compatibility with a recent kernel ABI breakage.
143 * /etc/rc.local is now hooked in via a generator binary, and
144 thus will no longer act as synchronization point during
147 * systemctl list-unit-files now supports --root=.
149 * systemd-tmpfiles now understands two new commands: z, Z for
150 relabelling files according to the SELinux database. This is
151 useful to apply SELinux labels to specific files in /sys,
154 * Output of SysV services is now forwarded to both the console
155 and the journal by default, not only just the console.
157 * New man pages for all APIs from libsystemd-login.
159 * The build tree got reorganized and a the build system is a
160 lot more modular allowing embedded setups to specifically
161 select the components of systemd they are interested in.
163 * Support for Linux systems lacking the kernel VT subsystem is
166 * configure's --with-rootdir= got renamed to
167 --with-rootprefix= to follow the naming used by udev and
170 * Unless specified otherwise we'll now install to /usr instead
171 of /usr/local by default.
173 * Processes with '@' in argv[0][0] are now excluded from the
174 final shut-down killing spree, following the logic explained
176 http://www.freedesktop.org/wiki/Software/systemd/RootStorageDaemons
178 * All processes remaining in a service cgroup when we enter
179 the START or START_PRE states are now killed with
180 SIGKILL. That means it is no longer possible to spawn
181 background processes from ExecStart= lines (which was never
182 supported anyway, and bad style).
184 * New PropagateReloadTo=/PropagateReloadFrom= options to bind
185 reloading of units together.
187 Contributions from: Bill Nottingham, Daniel Walsh, Dave
188 Reisner, Dexter Morgan, Gregs Gregs, Jonathan Nieder, Kay
189 Sievers, Lennart Poettering, Michael Biebl, Michal Schmidt,
190 Michał Górny, Ran Benita, Thomas Jarosch, Tim Waugh, Tollef
191 Fog Heen, Tom Gundersen, Zbigniew Jędrzejewski-Szmek