]>
err.no Git - systemd/log
Lennart Poettering [Mon, 4 Apr 2011 16:15:13 +0000 (18:15 +0200)]
exec: support unlimited resources
Lennart Poettering [Mon, 4 Apr 2011 14:56:51 +0000 (16:56 +0200)]
selinux: relabel /run the same way as /dev after loading the policy since they both come pre-filled and unlabelled
Lennart Poettering [Mon, 4 Apr 2011 13:38:25 +0000 (15:38 +0200)]
update TODO
Michal Schmidt [Mon, 4 Apr 2011 13:08:40 +0000 (15:08 +0200)]
manager: fd must be int, not char
This should fix the crash reported by Dan Horak on s390x which
does not have VTs.
Kay Sievers [Mon, 4 Apr 2011 13:33:00 +0000 (15:33 +0200)]
change remaining /var/run to /run
Lennart Poettering [Mon, 4 Apr 2011 13:22:58 +0000 (15:22 +0200)]
units: move user units from /usr/share to /usr/lib since they might be arch-dependent
Lennart Poettering [Mon, 4 Apr 2011 13:23:29 +0000 (15:23 +0200)]
logger: name socket like service
Lennart Poettering [Mon, 4 Apr 2011 02:06:24 +0000 (04:06 +0200)]
units: improve logger descriptions
Lennart Poettering [Mon, 4 Apr 2011 01:48:09 +0000 (03:48 +0200)]
man: document /etc/sysctl.d/
Lennart Poettering [Mon, 4 Apr 2011 01:36:42 +0000 (03:36 +0200)]
binfmt: add binfmt tool to set up binfmt_misc at boot
Lennart Poettering [Sun, 3 Apr 2011 21:55:40 +0000 (23:55 +0200)]
update TODO
Kay Sievers [Sun, 3 Apr 2011 20:21:21 +0000 (22:21 +0200)]
tmpfiles: create leading directories for d/D instructions
Lennart Poettering [Sun, 3 Apr 2011 20:18:35 +0000 (22:18 +0200)]
condition: use 'selinux' rather than 'SELinux' as preferred spelling
The virtualization condition and others use lowercase identifiers, so
for the sake of keeping things least surprising, use lowercase
identifiers here too.
Michal Schmidt [Sun, 3 Apr 2011 16:17:05 +0000 (18:17 +0200)]
man: document ConditionSecurity
Michal Schmidt [Sun, 3 Apr 2011 16:16:59 +0000 (18:16 +0200)]
condition: add ConditionSecurity
Using ConditionSecurity a unit can depend on a security module being
enabled/disabled. For now the only recognized security module is SELinux.
I'd like to use this feature for a unit that creates /.autorelabel if
SELinux is disabled, to ensure a relabel is done automatically when the
system is later rebooted with SELinux enabled.
Michal Schmidt [Sun, 3 Apr 2011 16:16:54 +0000 (18:16 +0200)]
load-fragment: unify config_parse_condition_{kernel, virt}
They only differ in the condition type, otherwise the code is identical.
Replace them with a more generic config_parse_condition_string().
Michal Schmidt [Sun, 3 Apr 2011 16:16:48 +0000 (18:16 +0200)]
condition: fix dumping of conditions
Several condition types were missing their strings, they were showing as
"(null)" in systemctl dump.
Indentation was missing too.
Lennart Poettering [Sun, 3 Apr 2011 20:14:56 +0000 (22:14 +0200)]
initctl: /dev/initctl is a named pipe, not a socket
Lennart Poettering [Sun, 3 Apr 2011 20:14:34 +0000 (22:14 +0200)]
kmsg-syslogd: pass facility value into kmsg
Kay Sievers [Sun, 3 Apr 2011 20:09:25 +0000 (22:09 +0200)]
move /var/lock to HAVE_SYSV_COMPAT
Kay Sievers [Sun, 3 Apr 2011 15:08:46 +0000 (17:08 +0200)]
update TODO
Lennart Poettering [Fri, 1 Apr 2011 23:08:31 +0000 (01:08 +0200)]
tmpfiles: split off rules for legacy systems into legacy.conf
Lennart Poettering [Fri, 1 Apr 2011 13:25:46 +0000 (15:25 +0200)]
general: replace a few uses of /var/run by /run
Lennart Poettering [Thu, 31 Mar 2011 23:35:27 +0000 (01:35 +0200)]
build-sys: bump version
Lennart Poettering [Thu, 31 Mar 2011 22:43:28 +0000 (00:43 +0200)]
tmpfiles: enforce new /var/lock semantics
http://lists.freedesktop.org/archives/systemd-devel/2011-March/001823.html
Lennart Poettering [Thu, 31 Mar 2011 22:42:57 +0000 (00:42 +0200)]
man: document ConditionPathIsDirectory=
Lennart Poettering [Thu, 31 Mar 2011 19:40:10 +0000 (21:40 +0200)]
mount: also relabel pre-mounted API dirs
Lennart Poettering [Thu, 31 Mar 2011 19:22:44 +0000 (21:22 +0200)]
log: fix shifting of facilities
Lennart Poettering [Thu, 31 Mar 2011 17:49:04 +0000 (19:49 +0200)]
log: don't strip facility when writing to kmsg
Lennart Poettering [Thu, 31 Mar 2011 16:19:59 +0000 (18:19 +0200)]
man: uinput.ko is a bad example
Lennart Poettering [Thu, 31 Mar 2011 16:17:29 +0000 (18:17 +0200)]
build-sys: create a number of drop-in config dirs
Lennart Poettering [Thu, 31 Mar 2011 13:35:40 +0000 (15:35 +0200)]
src: our lord is coverity
Lennart Poettering [Thu, 31 Mar 2011 02:40:02 +0000 (04:40 +0200)]
random: do not print warning if random seed doesn't exist yet
Lennart Poettering [Thu, 31 Mar 2011 02:26:40 +0000 (04:26 +0200)]
plymouth: use PID file to detect whether ply is running
Lennart Poettering [Thu, 31 Mar 2011 01:25:30 +0000 (03:25 +0200)]
build-sys: install systemd-analyze by default
Lennart Poettering [Thu, 31 Mar 2011 01:17:13 +0000 (03:17 +0200)]
analyze: improve output
Kay Sievers [Thu, 31 Mar 2011 00:36:32 +0000 (02:36 +0200)]
update TODO
Lennart Poettering [Wed, 30 Mar 2011 23:19:12 +0000 (01:19 +0200)]
analyze: add plotter
Lennart Poettering [Wed, 30 Mar 2011 18:16:07 +0000 (20:16 +0200)]
unit: when deserializing do reconnect to dbus/syslog when they show up
Lennart Poettering [Wed, 30 Mar 2011 18:15:45 +0000 (20:15 +0200)]
analyze: beautify output a bit
Lennart Poettering [Wed, 30 Mar 2011 18:04:37 +0000 (20:04 +0200)]
analyze: add systemd-analyze tool
Lennart Poettering [Wed, 30 Mar 2011 18:04:20 +0000 (20:04 +0200)]
unit: don't override timestamps due to state changes when deserializing
Lennart Poettering [Wed, 30 Mar 2011 01:03:35 +0000 (03:03 +0200)]
plymouth: don't explicitly enable status message when plymouth is up
https://bugzilla.redhat.com/show_bug.cgi?id=676302
systemd now watches /run/initramfs/plymouth and generates messages
exactly when that file exists. Hence we don't need the sending of the
signals anymore.
Lennart Poettering [Wed, 30 Mar 2011 00:21:48 +0000 (02:21 +0200)]
status: show status messages unconditionally if plymouth is around
Lennart Poettering [Wed, 30 Mar 2011 00:12:46 +0000 (02:12 +0200)]
taint: add missing cgroups taint flag
Lennart Poettering [Wed, 30 Mar 2011 00:09:15 +0000 (02:09 +0200)]
locale: don't access misinitialized variable
Lennart Poettering [Tue, 29 Mar 2011 23:53:34 +0000 (01:53 +0200)]
quota: do not pull in quota tools for mounts that do not originate in neither /etc/fstab nor fragment files
Lennart Poettering [Tue, 29 Mar 2011 22:47:50 +0000 (00:47 +0200)]
manager: fix taint check for /usr
Lennart Poettering [Tue, 29 Mar 2011 22:43:16 +0000 (00:43 +0200)]
unit: never apply /etc/rcN.d/ priority to native services
Lennart Poettering [Tue, 29 Mar 2011 21:32:31 +0000 (23:32 +0200)]
unit: fix parsing of condition-result
Lennart Poettering [Tue, 29 Mar 2011 21:32:10 +0000 (23:32 +0200)]
unit: don't complain about failed units when deserializing
Lennart Poettering [Tue, 29 Mar 2011 21:31:38 +0000 (23:31 +0200)]
exec: drop process group kill mode since it has little use and confuses the user
Lennart Poettering [Tue, 29 Mar 2011 18:29:02 +0000 (20:29 +0200)]
cgroup: explain when we cannot initialize the cgroup stuff
Florian Kriener [Tue, 29 Mar 2011 16:31:38 +0000 (18:31 +0200)]
locale: full fledged /etc/default/locale support for debian
debian uses /etc/default/locale for all locale setup, this patch
adds full support for it to systemd.
Lennart Poettering [Tue, 29 Mar 2011 16:32:46 +0000 (18:32 +0200)]
systemctl: don't truncate description when using pager
https://bugs.freedesktop.org/show_bug.cgi?id=35725
Lennart Poettering [Tue, 29 Mar 2011 16:32:46 +0000 (18:32 +0200)]
ask-password: also accept Backspace as first keypress as silent mode switch
Lennart Poettering [Tue, 29 Mar 2011 11:11:13 +0000 (13:11 +0200)]
unit: when deserializing jobs, don't pull in dependencies
Lennart Poettering [Tue, 29 Mar 2011 10:11:14 +0000 (12:11 +0200)]
man: really fix all LC_MESSAGE to LC_MESSAGES
https://bugs.freedesktop.org/show_bug.cgi?id=35534
Lennart Poettering [Tue, 29 Mar 2011 00:20:05 +0000 (02:20 +0200)]
build-sys: bump version
Kay Sievers [Mon, 28 Mar 2011 23:41:50 +0000 (01:41 +0200)]
udev: tty - re-add accidentially removed hvc* match
Lennart Poettering [Mon, 28 Mar 2011 23:38:06 +0000 (01:38 +0200)]
locale: fix LC_MESSAGES variable name
https://bugs.freedesktop.org/show_bug.cgi?id=35534
William Jon McCann [Sun, 27 Mar 2011 04:18:16 +0000 (00:18 -0400)]
plymouth: Remove the calls to plymouth message
* messages aren't translated
* console text is ugly
* they are jargonny
* they really aren't needed
https://bugs.freedesktop.org/show_bug.cgi?id=35711
Kay Sievers [Mon, 28 Mar 2011 23:19:39 +0000 (01:19 +0200)]
udev: systemd-tag all ttys
Kay Sievers [Mon, 28 Mar 2011 22:15:14 +0000 (00:15 +0200)]
tmpfiles fix /run/lock permissions
<mbiebl> kay: just wondering: d /run/lock 0755 root lock -
<mbiebl> shouldn't that rather be 0775?
<mbiebl> otherwise it doesn't make sense
Lennart Poettering [Mon, 28 Mar 2011 21:39:18 +0000 (23:39 +0200)]
man: fix references to systemd.unit= on the kernel cmdline
https://bugs.freedesktop.org/show_bug.cgi?id=35720
Michael Olbrich [Mon, 21 Mar 2011 16:16:50 +0000 (17:16 +0100)]
udev: expose some more tty's for various embedded hardware
Lennart Poettering [Mon, 28 Mar 2011 21:27:04 +0000 (23:27 +0200)]
ask-password: use TAB instead of backspace to disable asterisk password echo
Jan Engelhardt [Sun, 27 Mar 2011 21:52:11 +0000 (23:52 +0200)]
crypto: to show stars or not to show them
On Friday 2011-03-18 01:41, Lennart Poettering wrote:
>On Fri, 18.03.11 00:18, Jan Engelhardt (jengelh@medozas.de) wrote:
>
>> Meanwhile, I have two new suggestions.
>
>I have one too (or actually Kay came up with it), and I think you are
>going to like it:
>
>Start with showing input feedback as we currently do. If the user then
>presses TAB the stars disappear, and instead we show "(no echo)" or
>so. Then, the user can proceed with typing his password without
>asterisks.
>[...]
Incorporating Graham's suggestion to use BKSP instead:
The following changes since commit
65c9e467528daa438167853cc91d37bfcb875836 :
tainted: don't check if /usr is a mount point, only if it's not already mounted at startup (2011-03-24 22:32:21 +0100)
are available in the git repository at:
git://dev.medozas.de/systemd master
Jan Engelhardt (1):
ask-password: provide a way to activate a silent prompt
src/ask-password-api.c | 11 +++++++----
1 files changed, 7 insertions(+), 4 deletions(-)
Lennart Poettering [Mon, 28 Mar 2011 21:04:30 +0000 (23:04 +0200)]
conf-parser: fix remaining parser functions according to new prototype
Lennart Poettering [Mon, 28 Mar 2011 21:04:07 +0000 (23:04 +0200)]
execute: socket isn't abstract anymore
Kay Sievers [Fri, 25 Mar 2011 04:07:20 +0000 (05:07 +0100)]
use /run instead of /dev/.run
Instead of the /dev/.run trick we have currently implemented, we decided
to move the early-boot runtime dir to /run.
An existing /var/run directory is bind-mounted to /run. If /var/run is
already a symlink, no action is taken.
An existing /var/lock directory is bind-mounted to /run/lock.
If /var/lock is already a symlink, no action is taken.
To implement the directory vs. symlink logic, we have a:
ConditionPathIsDirectory=
now, which is used in the mount units.
Skipped mount unit in case of symlink:
$ systemctl status var-run.mount
var-run.mount - Runtime Directory
Loaded: loaded (/lib/systemd/system/var-run.mount)
Active: inactive (dead)
start condition failed at Fri, 25 Mar 2011 04:51:41 +0100; 6min ago
Where: /var/run
What: /run
CGroup: name=systemd:/system/var-run.mount
The systemd rpm needs to make sure to add something like:
%pre
mkdir -p -m0755 /run >/dev/null 2>&1 || :
or it needs to be added to filesystem.rpm.
Udev -git already uses /run if that exists, and is writable at bootup.
Otherwise it falls back to the current /dev/.udev.
Dracut and plymouth need to be adopted to switch from /dev/.run to run
too.
Cheers,
Kay
cee1 [Fri, 18 Mar 2011 02:03:41 +0000 (10:03 +0800)]
util: detect page size runtime.
Some architectures support multiple machine types with diffenent
page sizes, and some machine types even support multiple
page sizes themselves.
Lennart Poettering [Mon, 28 Mar 2011 19:36:13 +0000 (21:36 +0200)]
man: explain a couple of default dependencies
Kay Sievers [Thu, 24 Mar 2011 21:32:21 +0000 (22:32 +0100)]
tainted: don't check if /usr is a mount point, only if it's not already mounted at startup
Kay Sievers [Wed, 23 Mar 2011 00:32:40 +0000 (01:32 +0100)]
update TODO
Andrey Borzenkov [Tue, 22 Mar 2011 17:31:10 +0000 (20:31 +0300)]
man: no keep-root in pam_systemd anymore
Kay Sievers [Mon, 21 Mar 2011 14:09:12 +0000 (15:09 +0100)]
update TODO
Andrey Borzenkov [Fri, 18 Mar 2011 16:32:49 +0000 (19:32 +0300)]
units: replace Names=dm.service with symlink for mandriva prefdm
Andrey Borzenkov [Fri, 18 Mar 2011 16:12:58 +0000 (19:12 +0300)]
mount: pull in quota services from local mountpoints with usr/grpquota options
Lennart Poettering [Fri, 18 Mar 2011 04:26:25 +0000 (05:26 +0100)]
service: pull in sysv facility targets from the sysv units, not the other way round
For an explanation see:
http://lists.freedesktop.org/archives/systemd-devel/2011-March/001692.html
Lennart Poettering [Fri, 18 Mar 2011 04:17:02 +0000 (05:17 +0100)]
units: pull in syslog.target from syslog.socket
Lennart Poettering [Fri, 18 Mar 2011 04:13:06 +0000 (05:13 +0100)]
units: don't ever pull in SysV targets from other SysV targets
Lennart Poettering [Fri, 18 Mar 2011 04:12:34 +0000 (05:12 +0100)]
units: document that some targets exists only for compat with SysV
Lennart Poettering [Fri, 18 Mar 2011 04:01:30 +0000 (05:01 +0100)]
units: we don't need to generate syslog.target with sed anymore
Lennart Poettering [Fri, 18 Mar 2011 03:59:05 +0000 (04:59 +0100)]
units: we no longer need m4 to build graphical.target or multi-user.taregt
Lennart Poettering [Fri, 18 Mar 2011 03:49:53 +0000 (04:49 +0100)]
man: document pidns containers
Lennart Poettering [Fri, 18 Mar 2011 03:49:38 +0000 (04:49 +0100)]
units: deemphesize Names= settings, and explain why nobody whould use them
Lennart Poettering [Fri, 18 Mar 2011 03:41:47 +0000 (04:41 +0100)]
units: on mandriva/fedora create single.service alias via symlink, not Names=
Lennart Poettering [Fri, 18 Mar 2011 03:37:31 +0000 (04:37 +0100)]
units: get rid of runlevel Names=, the symlinks in /lib/systemd/system are much more useful
Lennart Poettering [Fri, 18 Mar 2011 03:32:58 +0000 (04:32 +0100)]
units: get rid of empty units/suse/ subdir
Lennart Poettering [Fri, 18 Mar 2011 03:31:22 +0000 (04:31 +0100)]
syslog: rework syslog detection so that we need no compile-time option what the name of the syslog implementation is
Lennart Poettering [Fri, 18 Mar 2011 02:32:47 +0000 (03:32 +0100)]
man: document .requires/ directories
Lennart Poettering [Fri, 18 Mar 2011 02:32:33 +0000 (03:32 +0100)]
special: get rid of dbus.target
Lennart Poettering [Fri, 18 Mar 2011 02:13:15 +0000 (03:13 +0100)]
exec: properly apply capability bounding set, add inverted bounding sets
Andrey Borzenkov [Thu, 17 Mar 2011 20:22:49 +0000 (23:22 +0300)]
dbus: add service D-Bus property "Sockets"
Andrey Borzenkov [Thu, 17 Mar 2011 20:22:49 +0000 (23:22 +0300)]
dbus: consolidate service SysV conditionals
No need to define the same set of properties twice. While on it,
add FsckPassNo to introspection.
Kay Sievers [Thu, 17 Mar 2011 14:15:36 +0000 (15:15 +0100)]
update TODO
Kay Sievers [Thu, 17 Mar 2011 13:03:17 +0000 (14:03 +0100)]
update TODO
Lennart Poettering [Thu, 17 Mar 2011 03:36:19 +0000 (04:36 +0100)]
unit: serialize condition test results
cee1 [Thu, 17 Mar 2011 02:13:01 +0000 (10:13 +0800)]
read-ahead: Fix broken systemd-readahead-collect on mips.
This actually adjust the __NR_fanotify* system call numbers to proper
ones on mips(according userspace ABI).