chiark / gitweb /
Lennart Poettering [Tue, 25 Dec 2012 10:37:06 +0000 (11:37 +0100)]
TODO
Lennart Poettering [Tue, 25 Dec 2012 10:36:37 +0000 (11:36 +0100)]
polkit: add "imply" rules to a number of polkit actions
Marcel Holtmann [Fri, 28 Dec 2012 07:53:26 +0000 (23:53 -0800)]
hwdb: Add database for Bluetooth company identifiers
Kay Sievers [Mon, 31 Dec 2012 03:48:44 +0000 (04:48 +0100)]
udev: in addition to DEVMODE, honor DEVUID, DEVGID from the uevent
Tom Gundersen [Fri, 28 Dec 2012 23:21:28 +0000 (00:21 +0100)]
fstab-generator: fix typo
Tom Gundersen [Fri, 23 Nov 2012 02:41:13 +0000 (03:41 +0100)]
fstab-generator: generate new_root.mount in initrd
The configuration is taken from /proc/cmdline, aiming at emulating the
behavior of the kernel when no initramfs is used.
The supported options are: root=, rootfstype=, rootwait=, rootflags=,
ro, and rw. rootdelay= was dropped, as it is not really useful in a
systemd world, but could easily be added.
v2: fix comments by Lennart, and complain loudly if root= can not be found
Cc: Harald Hoyer <harald@redhat.com>
Cc: Dave Reisner <dreisner@archlinux.org>
Tom Gundersen [Wed, 21 Nov 2012 22:22:03 +0000 (23:22 +0100)]
systemctl: switch-root - read init= from /proc/cmdline
If the path to init is not specified as an argumnt to systemctl, but
init= is given on the kernel commandline, use that.
This means the initrd does not need glue code to parse the kernel
commandline before passing on init= to systemctl.
Cc: Harald Hoyer <harald@redhat.com>
Cc: Dave Reisner <dreisner@archlinux.org>
Matthew Monaco [Mon, 24 Dec 2012 13:25:58 +0000 (06:25 -0700)]
pam_systemd: new option for the session class
Lennart Poettering [Mon, 24 Dec 2012 18:12:22 +0000 (19:12 +0100)]
update TODO
Lennart Poettering [Mon, 24 Dec 2012 18:03:59 +0000 (19:03 +0100)]
hostnamed: make chassis type configurable via /etc/machine-info
For many usecases it is useful to store the chassis type somewhere, and
/etc/machine-info sounds like a good place. Ideally we could always
detect the chassis type from firmware, but frequently that's not
available and in many embedded devices probably entirely unrealistic.
This patch adds a configurable setting CHASSIS= to /etc/machine-info and
exposes this via hostnamectl/hostnamed. hostnamed will guess the chassis
type from DMI if nothing is set explicitly. I also added support for
detecting it from ACPI, which should be more useful as ACPI 5.0 actually
knows a "tablet" chassis type, which neither DMI nor previous ACPI
versions knew.
This also enables DMI-based and ACPI-based detection for non-x86 systems
as ACPI is apparently coming to ARM platforms soon.
I tried to minimize the vocabulary of chassis types understood and
added: desktop, laptop, server, tablet, handset. This is much less than
either APCI or DMI know. If we need more types later on we can easily
add them.
Lennart Poettering [Mon, 24 Dec 2012 12:01:00 +0000 (13:01 +0100)]
fstab-generator: properly detect bind mounts
This kinda undoes
a83cbaccd03c3f28e47e9330f4a22ff65ce4b561 and
1d634e21b453f3c80d7c6c4bd90a6b84e42a3d2a but corrects the original code
to compare the mount type with "bind" rather than the mount options.
Lennart Poettering [Sun, 23 Dec 2012 21:38:21 +0000 (22:38 +0100)]
update TODO
Lennart Poettering [Sun, 23 Dec 2012 21:12:01 +0000 (22:12 +0100)]
logind: don't allow suspending/hibernating if the kernel doesn't support it
Lennart Poettering [Sun, 23 Dec 2012 20:58:37 +0000 (21:58 +0100)]
util: if /sys mounted read-only we can't suspend/hibernate
Lennart Poettering [Sun, 23 Dec 2012 16:28:17 +0000 (17:28 +0100)]
util: when determining controlling tty of a process properly handle processes without
Lennart Poettering [Sun, 23 Dec 2012 21:32:48 +0000 (22:32 +0100)]
logind: add support for automatic suspend/hibernate/shutdown on idle
Lennart Poettering [Sun, 23 Dec 2012 21:31:17 +0000 (22:31 +0100)]
pam: properly handle SSH logins lacking the PAM tty field
Lennart Poettering [Sun, 23 Dec 2012 10:56:22 +0000 (11:56 +0100)]
update TODO
Lennart Poettering [Sun, 23 Dec 2012 10:56:13 +0000 (11:56 +0100)]
systemctl: don't show cgroup field for a unit if cgroup is empty
Colin Walters [Sun, 23 Dec 2012 15:18:21 +0000 (10:18 -0500)]
build-sys: Make libsystemd-shared link to libsystemd-daemon
Commit
f934051c4d17d167bd8e46dd3c70a745e26fd2eb broke the build
because it made libsystemd-shared call sd_listen_fds() which is
defined in libsystemd-daemon.
This is a bit of a contortion because libsystemd-shared.la is a
noinst_LTLIBRARY, but libtool should do the right thing here and emit
DT_NEEDED on libsystemd-daemon.so for things that consume
libsystemd-shared.la.
Kay Sievers [Sun, 23 Dec 2012 13:21:19 +0000 (14:21 +0100)]
hwdb: update
Lennart Poettering [Sun, 23 Dec 2012 10:27:20 +0000 (11:27 +0100)]
shared: move fdset.c from core/ to shared/ since it is used by more than the core now
Lennart Poettering [Sun, 23 Dec 2012 10:23:59 +0000 (11:23 +0100)]
journalctl: strip TABs and ANSI color sequences from log messages when displaying them
Lennart Poettering [Sat, 22 Dec 2012 21:17:37 +0000 (22:17 +0100)]
update TODO
Lennart Poettering [Sat, 22 Dec 2012 20:21:09 +0000 (21:21 +0100)]
nspawn: try to orderly shutdown container when receiving SIGTERM
Lennart Poettering [Sat, 22 Dec 2012 18:30:07 +0000 (19:30 +0100)]
socket: support socket activation of containers
Lennart Poettering [Sat, 22 Dec 2012 18:29:04 +0000 (19:29 +0100)]
nspawn: allow passing socket activation fds through nspawn
Lennart Poettering [Sat, 22 Dec 2012 18:28:19 +0000 (19:28 +0100)]
fdset: add calls for initializing fdset from socket activation fds
Lennart Poettering [Sat, 22 Dec 2012 18:25:10 +0000 (19:25 +0100)]
socket: properly serialize/desrialize mqueue fds
Lennart Poettering [Sat, 22 Dec 2012 12:11:17 +0000 (13:11 +0100)]
nspawn: allow nspawn to be invoked without tty
This allows invoking nspawn containers as systemd services, to create a
minimal, light-weight OS container solution for servers.
Lukas Nykryn [Wed, 19 Dec 2012 12:12:26 +0000 (13:12 +0100)]
udevd: add missing ':' to getopt_long 'e'
Parameter -e is set without additional argument in getopt
and this leads to segfault when calling 'systemd-udevd -e'.
Lennart Poettering [Fri, 21 Dec 2012 21:20:46 +0000 (22:20 +0100)]
update TODO
Dimitrios Apostolou [Wed, 19 Dec 2012 21:32:52 +0000 (22:32 +0100)]
socket: Too many incoming connections
Hello list,
some socket activated service gave me the error message you can see on
the subject, maybe systemd should be more verbose in that case.
Thanks,
Dimitris
Lennart Poettering [Wed, 19 Dec 2012 19:50:42 +0000 (20:50 +0100)]
update TODO
Daniel Wallace [Wed, 19 Dec 2012 17:28:38 +0000 (12:28 -0500)]
zsh: add hibernate and hybrid-sleep to zsh-completion
Daniel Wallace [Fri, 7 Dec 2012 05:01:15 +0000 (00:01 -0500)]
unit: add %U for uids in units
Lennart Poettering [Wed, 19 Dec 2012 14:27:27 +0000 (15:27 +0100)]
update TODO
Lukas Nykryn [Wed, 19 Dec 2012 13:38:53 +0000 (14:38 +0100)]
coredumpctl: check return of strndup
Dave Reisner [Wed, 19 Dec 2012 13:11:20 +0000 (08:11 -0500)]
bash-completion: update list of verbs
Add suspend, hibernate, hybrid-sleep
Mantas Mikulėnas [Mon, 17 Dec 2012 20:32:21 +0000 (22:32 +0200)]
logind: allow HybridSleep methods in default DBus context
Michal Sekletar [Thu, 13 Dec 2012 13:59:40 +0000 (14:59 +0100)]
socket-util: added check of return value
Michal Sekletar [Thu, 13 Dec 2012 13:59:39 +0000 (14:59 +0100)]
path-util: set pointer to null after calling free()
In cases where path_strv_canonicalize() returns NULL, strv_free() is
called afterwards and it will call free() on pointers which were freed
already in path_strv_canonicalize()
Sami Kerola [Mon, 17 Dec 2012 10:13:17 +0000 (11:13 +0100)]
sd-daemon: fix cppcheck warnings
Found with 'cppcheck --enable=all --inconclusive --std=posix' while
working with util-linux, which has a copy of this file.
[misc-utils/sd-daemon.c:363]: (style) Checking if unsigned variable \
'length' is less than zero.
[misc-utils/sd-daemon.c:366]: (style) Checking if unsigned variable \
'length' is less than zero.
References: http://www.spinics.net/lists/util-linux-ng/msg07031.html
Sami Kerola [Sun, 16 Dec 2012 22:23:23 +0000 (22:23 +0000)]
licence: remove references to old FSF address
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=57206
Zbigniew Jędrzejewski-Szmek [Mon, 17 Dec 2012 10:38:47 +0000 (11:38 +0100)]
journal: move _packed_ attribute to the end
GCC manual states that "For an enum, struct or union type, you may
specify attributes either between the enum, struct or union tag and
the name of the type, or just past the closing curly brace of the
definition. The former syntax is preferred." This means that the
attribute should not be located before 'struct'. Putting it between
'struct' and the name seems cluttered. Putting it at the end seems
most readable.
This avoids clang warnings.
Richard Yao [Sun, 18 Nov 2012 19:55:10 +0000 (14:55 -0500)]
Fix "attribute 'packed' is ignored" warnings when using Clang
Clang 3.1 warned that "attribute 'packed' is ignored". This stems from
placing "__attribute__ ((packed))" at the start of structure
declarations when common practice is to place it at the end.
Signed-off-by: Richard Yao <ryao@gentoo.org>
Kay Sievers [Fri, 14 Dec 2012 17:23:36 +0000 (18:23 +0100)]
hwdb: update comments and data
Michael Terry [Fri, 14 Dec 2012 14:02:13 +0000 (09:02 -0500)]
libudev: validate 'udev' argument to udev_enumerate_new()
https://bugs.freedesktop.org/show_bug.cgi?id=58289
Tollef Fog Heen [Fri, 14 Dec 2012 08:17:14 +0000 (09:17 +0100)]
Remove frugalware display-manager.service
Frugalware are shipping their own .service file now, so stop shipping
it here.
This effectively reverts
196e3fa74a88a04b0ecec7d3af648287dd088f8a
Michal Schmidt [Wed, 12 Dec 2012 21:24:04 +0000 (22:24 +0100)]
localectl: fix dbus call arguments in set_x11_keymap
Fixes an assertion failure in the dbus lib.
https://bugzilla.redhat.com/show_bug.cgi?id=882212
Eric Benoit [Tue, 11 Dec 2012 14:32:39 +0000 (15:32 +0100)]
libudev: hwdb: add missing endianess conversion
https://bugs.freedesktop.org/show_bug.cgi?id=58067
Kay Sievers [Sun, 9 Dec 2012 13:49:28 +0000 (14:49 +0100)]
udev: net_id: update comments
Kay Sievers [Sun, 9 Dec 2012 13:48:55 +0000 (14:48 +0100)]
hwdb: update
Michal Schmidt [Fri, 7 Dec 2012 17:02:43 +0000 (18:02 +0100)]
shutdown, umount: use verbs consistently
Mounts are "unmounted".
Swaps are "deactivated", not "turned off" nor "disabled".
Loop and DM devices are "detached", not "deleted".
Especially the deleting sounded a bit scary.
Michal Schmidt [Fri, 7 Dec 2012 16:44:50 +0000 (17:44 +0100)]
shutdown, umount: logging improvements
In bugreports about hangs during the late shutdown we are often missing
important information - what were we trying to unmount/detach when it hung.
Instead of printing what we successfully unmounted, print what we are
going to unmount/detach. And add messages to mark the completion of
categories (mount/swap/loop/DM).
Michal Schmidt [Fri, 7 Dec 2012 16:34:21 +0000 (17:34 +0100)]
umount: fix check for DM changed
delete_dm() returns 0 on success. The check for "r > 0" was likely
a copy&paste error from the loopback code where "r > 0" makes sense.
Michal Schmidt [Fri, 7 Dec 2012 16:28:30 +0000 (17:28 +0100)]
shutdown: downgrade a warning
All messages of the kind "not all done, %d left" are log_info, except
the one for DM devices. Make it info too.
Kay Sievers [Thu, 6 Dec 2012 21:46:22 +0000 (22:46 +0100)]
hwdb: update
Mantas Mikulėnas [Thu, 6 Dec 2012 21:34:25 +0000 (23:34 +0200)]
hwdb: ids-update: close the right file handles
The script was closing *INP and *OUTP, which never actually existed.
Also fix pci_vendor() opening usb.ids and immediately discarding the fh.
Kay Sievers [Thu, 6 Dec 2012 20:48:49 +0000 (21:48 +0100)]
rules: usb-hid-pm - remove AMI devices
On Thu, Dec 6, 2012 at 9:20 AM, Robert Milasan <rmilasan@suse.com> wrote:
> It seems that the added rules:
>
> ACTION=="add", SUBSYSTEM=="usb", ATTR{idVendor}=="046b",
> ATTR{idProduct}=="ff10", TEST=="power/control",
> ATTR{power/control}="auto"
>
> creates problems for people with Supermicro X8ST3 mb (and maybe
> other Supermicro mb's) and renders the KVM-over-IP unusable, at BIOS
> and GRUB the KVM works perfectly, but after that the device is unusable.
>
> Dropping the rule fixes the issue.
>
> Reference bug: https://bugzilla.novell.com/show_bug.cgi?id=792576
Kay Sievers [Thu, 6 Dec 2012 20:37:10 +0000 (21:37 +0100)]
udev: net_id - fix OUI handling
Kay Sievers [Thu, 6 Dec 2012 20:27:42 +0000 (21:27 +0100)]
udev: net_id - fix typos
Kay Sievers [Thu, 6 Dec 2012 20:25:39 +0000 (21:25 +0100)]
udev: net_id - update examples
Kay Sievers [Thu, 6 Dec 2012 20:09:30 +0000 (21:09 +0100)]
udev: net_id - update examples
Kay Sievers [Thu, 6 Dec 2012 19:42:49 +0000 (20:42 +0100)]
udev: net_id - provide path names for USB devices
Michal Schmidt [Thu, 6 Dec 2012 15:20:10 +0000 (16:20 +0100)]
systemctl: verbose message on missing [Install]
People still don't understand what the message implies.
We have to be more verbose (or more intelligent and detect some of the
cases automatically, but that's not so easy).
https://bugzilla.redhat.com/show_bug.cgi?id=884438
Zbigniew Jędrzejewski-Szmek [Wed, 5 Dec 2012 10:59:05 +0000 (11:59 +0100)]
core: add catalog entry and MESSAGE_ID for overmounting
Kay Sievers [Wed, 5 Dec 2012 01:08:49 +0000 (02:08 +0100)]
build-sys: add libsystemd-daemon.la to libsystemd-core.la
https://bugs.freedesktop.org/show_bug.cgi?id=57900
Philip Nilsson [Mon, 3 Dec 2012 17:26:52 +0000 (18:26 +0100)]
man: correct documented return type for sd_journal_close
Zbigniew Jędrzejewski-Szmek [Tue, 4 Dec 2012 12:57:53 +0000 (13:57 +0100)]
modules-load: fix memory leak when bailing out
Suggested-by: Mauro Dreissig <mukadr@gmail.com>
Tom Gundersen [Mon, 3 Dec 2012 19:15:22 +0000 (20:15 +0100)]
hwdb: add IAB to the OUI database
The individual address block is a poor man's organizationally unique
identifier.
Perhaps we should change the udev key from ID_OUI_FROM_DATABASE to
something like ID_IEEE_VENDOR_FROM_DATABASE?
Suggested-by: Diego Elio Pettenò <flameeyes@flameeyes.eu>
Umut Tezduyar [Mon, 3 Dec 2012 14:45:53 +0000 (15:45 +0100)]
systemctl: fix typos in help text
Kay Sievers [Mon, 3 Dec 2012 14:47:18 +0000 (15:47 +0100)]
udev: net_id - clarify comments
Shawn Landden [Mon, 3 Dec 2012 00:50:55 +0000 (00:50 +0000)]
timedated: do not incorrectly close non-opened dbus connection
Fix the fallowing error when no system dbus available:
Failed to get system D-Bus connection: Failed to connect to socket /var/run/dbus/system_bus_socket: No such file or directory
process 14920: arguments to dbus_connection_close() were incorrect, assertion "connection != NULL" failed in file ../../dbus/dbus-connection.c line 2889.
This is normally a bug in some application using the D-Bus library.
process 14920: arguments to dbus_connection_unref() were incorrect, assertion "connection != NULL" failed in file ../../dbus/dbus-connection.c line 2776.
This is normally a bug in some application using the D-Bus library.
David Herrmann [Sun, 2 Dec 2012 17:06:35 +0000 (18:06 +0100)]
getty-generator: fix handling systems without a console
If the system does not have any active console, we should not try to
create an empty symlink. Instead, create no symlink at all.
Otherwise, on systems with CONFIG_VT=n and no serial console, we will
create a symlink with an empty template parameter.
Kay Sievers [Sun, 2 Dec 2012 14:16:20 +0000 (15:16 +0100)]
hwdb: fix upstream typo in PNP data
Kay Sievers [Sun, 2 Dec 2012 13:58:47 +0000 (14:58 +0100)]
hwdb: update PNP ids
David Strauss [Sun, 2 Dec 2012 03:48:42 +0000 (19:48 -0800)]
Grammar and spelling fixes to journal documentation.
Kay Sievers [Sun, 2 Dec 2012 01:10:40 +0000 (02:10 +0100)]
gitignore: ignore .trs files from tests
Kay Sievers [Sun, 2 Dec 2012 01:01:25 +0000 (02:01 +0100)]
hwdb: update data
Kay Sievers [Sun, 2 Dec 2012 01:00:57 +0000 (02:00 +0100)]
udev: hwdb - fix OUI lookup
Zbigniew Jędrzejewski-Szmek [Thu, 29 Nov 2012 19:26:12 +0000 (20:26 +0100)]
man: fix docbook in systemd-coredumpctl(1)
Zbigniew Jędrzejewski-Szmek [Wed, 28 Nov 2012 16:47:10 +0000 (17:47 +0100)]
man: document more URLs for journal-gatewayd
Eelco Dolstra [Fri, 30 Nov 2012 12:36:05 +0000 (13:36 +0100)]
man: fix some typos
Zbigniew Jędrzejewski-Szmek [Sat, 1 Dec 2012 09:37:52 +0000 (10:37 +0100)]
build-sys: use parallel tests driver
This saves test output to individual .log files.
The driver is only used in /Makefile.am, not in
/docs/*udev/Makefile.am because the latter don't seem to work with
this driver. They don't produce much output anyway.
.gitignore is alphabetized, and .log files are added to it.
Generated files from /build-aux are removed from the list.
Zbigniew Jędrzejewski-Szmek [Sat, 1 Dec 2012 09:13:30 +0000 (10:13 +0100)]
build-sys: color test results
Zbigniew Jędrzejewski-Szmek [Mon, 19 Nov 2012 19:19:39 +0000 (20:19 +0100)]
cgls: automatic cleanup
Kay Sievers [Fri, 30 Nov 2012 18:27:42 +0000 (19:27 +0100)]
udev: net_id - suppress function number for single-function devices
Ramkumar Ramachandra [Thu, 29 Nov 2012 19:06:57 +0000 (00:36 +0530)]
man: fix some spelling mistakes
Ramkumar Ramachandra [Thu, 29 Nov 2012 19:06:56 +0000 (00:36 +0530)]
DISTRO_PORTING: correct location of hostname-setup.c
f33d3ec1 (move more common files to shared/ and add them to shared.la,
2012-04-12) moved src/hostname-setup.c to src/shared/hostname-setup.c,
but did not update DISTRO_PORTING accordingly. Fix this.
Ramkumar Ramachandra [Thu, 29 Nov 2012 19:06:55 +0000 (00:36 +0530)]
LICENSE.LGPL2.1: update license
Update the license with the copy found on
http://www.gnu.org/licenses/old-licenses/lgpl-2.1.txt
Martin Pitt [Wed, 28 Nov 2012 20:33:26 +0000 (21:33 +0100)]
keymap: Add Samsung 700T
https://bugzilla.redhat.com/show_bug.cgi?id=879691
Alessandro Crismani [Wed, 8 Aug 2012 12:19:32 +0000 (14:19 +0200)]
keymap: Update the list of Samsung Series 9 models
Currently, keymaps are provided only for the NP90X3A laptop. Samsung
introduced updated models, codenamed 900X3B, 900X3C, 900X4B, 900X4C,
which are currently not matched by udev rules. This patch includes the
newer modules in udev rules and move the samsung-n90x3a file defining
keys to a more generic samsung-series-9 file.
The patch was tested on a 900X4C laptop, and other people reported
that the rules also work for 900X3B and 900X3C ones.
Hermann Gausterer [Wed, 28 Nov 2012 20:20:37 +0000 (21:20 +0100)]
shutdown: fix typo
Tom Gundersen [Wed, 28 Nov 2012 17:08:54 +0000 (18:08 +0100)]
shutdown: don't consider umounting of / and /usr failed
In the words of Homer: If you don't try, you can't fail.
This is a revert of
9279749b84cc87c7830280b7895a48bed03c9429.
It used to be necessary to consider the umounting failed to make sure /
and /usr were remounted read-only, but that is no longer necessary as
everything is now remounted read-only anyway.
Moreover, this avoids a warning at shutdown saying a filesystem was not
unmounted. As the umounting of / is never attempted there was no
corresponding warning message saying which fs that failed. This caused some
spurious bug-reports from concerned users.
Cc: Michael Biebl <biebl@debian.org>
Tom Gundersen [Wed, 28 Nov 2012 16:27:17 +0000 (17:27 +0100)]
shutdown: don't pivot to /run/initramfs if already there
This allows systemd-shutdown to be installed as /run/initramfs/shutdown
and take care of the proper teardown of the rootfs.
Kay Sievers [Tue, 27 Nov 2012 10:40:08 +0000 (11:40 +0100)]
man: systemd.exec - mention mount(2)
https://bugzilla.redhat.com/show_bug.cgi?id=880552
Tom Gundersen [Tue, 27 Nov 2012 00:09:28 +0000 (01:09 +0100)]
fstab-generator: make error more helpful in case of duplicates in fstab
Traditional sysvinit systems would not complain about duplicates in
fstab. Rather it (through monut -a) would mount one fs on top of another,
in effect the last entry taking precedent.
In systemd, the first entry takes precedent, all subsequent ones are
ignored and an error is printed.
The change of behavior and the source of this error message was causing
some confusion, so give a hint what migt be wrong.
Tom Gundersen [Mon, 26 Nov 2012 23:52:01 +0000 (00:52 +0100)]
fstab-generator: drop mount_is_bind
Do the check in-line instead, as the function is now a bit pointless.
Tom Gundersen [Mon, 26 Nov 2012 02:49:21 +0000 (03:49 +0100)]
fstab-generator: remove redundancy in mount_is_bind