]> git.ipfire.org Git - thirdparty/systemd.git/blame - NEWS
modules-load: resolve aliases before insert modules
[thirdparty/systemd.git] / NEWS
CommitLineData
d657c51f 1systemd System and Service Manager
220a21d3 2
e0d25329
KS
3CHANGES WITH 41:
4 * The systemd binary is installed /usr/lib/systemd/systemd now;
5 An existing /sbin/init symlink needs to be adapted with the
6 package update.
7
b13df964
LP
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.
11
12 * Watchdog support is now already useful, but still not
13 complete.
14
15 * A new kernel command line option systemd.setenv= is
16 understood to set system wide environment variables
17 dynamically at boot.
18
19 Contributions from: Benjamin Franzke, Kay Sievers, Lennart
20 Poettering, Michael Olbrich, Michal Schmidt, Tom Gundersen,
21 William Douglas
22
d26e4270
LP
23CHANGES WITH 40:
24 * This is mostly a bugfix release
25
26 * We now expose the reason why a service failed in the
27 "Result" D-Bus property.
28
29 * Rudimentary service watchdog support (will be completed over
30 the next few releases.)
31
32 * When systemd forks off in order execute some service we will
33 now immediately changes its argv[0] to reflect which process
34 it will execute. This is useful to minimize the time window
35 with a generic argv[0], which makes bootcharts more useful
36
b13df964
LP
37 Contributions from: Alvaro Soliverez, Chris Paulson-Ellis, Kay
38 Sievers, Lennart Poettering, Michael Olbrich, Michal Schmidt,
39 Mike Kazantsev, Ray Strode
40
220a21d3
LP
41CHANGES WITH 39:
42 * This is mostly a test release, but incorporates many
43 bugfixes.
44
45 * New systemd-cgtop tool to show control groups by their
46 resource usage.
47
48 * Linking against libacl for ACLs is optional again. If
49 disabled, support tracking device access for active logins
50 goes becomes unavailable, and so does access to the user
51 journals by the respective users.
52
53 * If a group "adm" exists, journal files are automatically
54 owned by them, thus allow members of this group full access
55 to the system journal as well as all user journals.
56
57 * The journal now stores the SELinux context of the logging
58 client for all entries.
59
60 * Add C++ inclusion guards to all public headers
61
62 * New output mode "cat" in the journal to print only text
63 messages, without any meta data like date or time.
64
65 * Include tiny X server wrapper as a temporary stop-gap to
66 teach XOrg udev display enumeration. This is used by display
67 managers such as gdm, and will go away as soon as XOrg
68 learned native udev hotplugging for display devices.
69
70 * Add new systemd-cat tool for executing arbitrary programs
71 with STDERR/STDOUT connected to the journal. Can also act as
72 BSD logger replacement, and does so by default.
73
74 * Optionally store all locally generated coredumps in the
75 journal along with meta data.
76
77 * systemd-tmpfiles learnt four new commands: n, L, c, b, for
78 writing short strings to files (for usage for /sys), and for
79 creating symlinks, character and block device nodes.
80
81 * New unit file option ControlGroupPersistent= to make cgroups
82 persistent, following the mechanisms outlined in
83 http://www.freedesktop.org/wiki/Software/systemd/PaxControlGroups
84
85 * Support multiple local RTCs in a sane way
86
87 * No longer monopolize IO when replaying readahead data on
88 rotating disks, since we might starve non-file-system IO to
89 death, since fanotify() will not see accesses done by blkid,
90 or fsck.
91
92 * Don't show kernel threads in systemd-cgls anymore, unless
93 requested with new -k switch.
94
95 Contributions from: Dan Horák, Kay Sievers, Lennart
96 Poettering, Michal Schmidt
97
98CHANGES WITH 38:
99 * This is mostly a test release, but incorporates many
100 bugfixes.
101
102 * The git repository moved to:
103 git://anongit.freedesktop.org/systemd/systemd
104 ssh://git.freedesktop.org/git/systemd/systemd
105
106 * First release with the journal
107 http://0pointer.de/blog/projects/the-journal.html
108
109 * The journal replaces both systemd-kmsg-syslogd and
110 systemd-stdout-bridge.
111
112 * New sd_pid_get_unit() API call in libsystemd-logind
113
114 * Many systemadm clean-ups
115
116 * Introduce remote-fs-pre.target which is ordered before all
117 remote mounts and may be used to start services before all
118 remote mounts.
119
120 * Added Mageia support
121
122 * Add bash completion for systemd-loginctl
123
124 * Actively monitor PID file creation for daemons which exit in
125 the parent process before having finished writing the PID
126 file in the daemon process. Daemons which do this need to be
127 fixed (i.e. PID file creation must have finished before the
128 parent exits), but we now react a bit more gracefully to them.
129
130 * Add colourful boot output, mimicking the well-known output
131 of existing distributions.
132
133 * New option PassCredentials= for socket units, for
134 compatibility with a recent kernel ABI breakage.
135
136 * /etc/rc.local is now hooked in via a generator binary, and
137 thus will no longer act as synchronization point during
138 boot.
139
140 * systemctl list-unit-files now supports --root=.
141
142 * systemd-tmpfiles now understands two new commands: z, Z for
143 relabelling files according to the SELinux database. This is
144 useful to apply SELinux labels to specific files in /sys,
145 among other things.
146
147 * Output of SysV services is now forwarded to both the console
148 and the journal by default, not only just the console.
149
150 * New man pages for all APIs from libsystemd-login.
151
152 * The build tree got reorganized and a the build system is a
153 lot more modular allowing embedded setups to specifically
154 select the components of systemd they are interested in.
155
156 * Support for Linux systems lacking the kernel VT subsystem is
157 restored.
158
159 * configure's --with-rootdir= got renamed to
160 --with-rootprefix= to follow the naming used by udev and
161 kmod
162
163 * Unless specified otherwise we'll now install to /usr instead
164 of /usr/local by default.
165
166 * Processes with '@' in argv[0][0] are now excluded from the
167 final shut-down killing spree, following the logic explained
168 in:
169 http://www.freedesktop.org/wiki/Software/systemd/RootStorageDaemons
170
171 * All processes remaining in a service cgroup when we enter
172 the START or START_PRE states are now killed with
173 SIGKILL. That means it is no longer possible to spawn
174 background processes from ExecStart= lines (which was never
175 supported anyway, and bad style).
176
177 * New PropagateReloadTo=/PropagateReloadFrom= options to bind
178 reloading of units together.
179
180 Contributions from: Bill Nottingham, Daniel Walsh, Dave
181 Reisner, Dexter Morgan, Gregs Gregs, Jonathan Nieder, Kay
182 Sievers, Lennart Poettering, Michael Biebl, Michal Schmidt,
183 Michał Górny, Ran Benita, Thomas Jarosch, Tim Waugh, Tollef
184 Fog Heen, Tom Gundersen, Zbigniew Jędrzejewski-Szmek