chiark / gitweb /
elogind.git
4 years agorules: simplify mmc RPMB handling
Martin Pitt [Fri, 13 Feb 2015 09:59:38 +0000 (10:59 +0100)]
rules: simplify mmc RPMB handling

We don't actually want a by-path/ symlink for MMC RPMB devices, so just add
them to the blacklist. This will prevent creating wrong by-path links and
blkid'ing those.

4 years agorules: Fix by-path of mmc RPMB partitions and don't blkid them
Martin Pitt [Wed, 11 Feb 2015 14:26:52 +0000 (15:26 +0100)]
rules: Fix by-path of mmc RPMB partitions and don't blkid them

Linux 3.10+ exposes RPMB (Replay Protected Memory Block) partitions of MMC
devices [1] ; trying to read them with blkid or other unspecific means will
cause kernel buffer I/O errors and timeouts. So don't run blkid on these.

Also ensure that /dev/disk/by-path creates proper symlinks and exposes the
-rpmb partition separately, instead of letting the "normal" partition symlink
point to the rpbm device (this is a race condition).

[1] http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=090d25fe224c0

https://launchpad.net/bugs/1333140

4 years agoinclude <poll.h> instead of <sys/poll.h>
Thomas Hindoe Paaboel Andersen [Thu, 12 Feb 2015 13:06:32 +0000 (14:06 +0100)]
include <poll.h> instead of <sys/poll.h>

include-what-you-use automatically does this and it makes finding
unnecessary harder to spot. The only content of poll.h is a include
of sys/poll.h so should be harmless.

4 years agoAdd missing includes in header files
Thomas Hindoe Paaboel Andersen [Tue, 10 Feb 2015 11:56:53 +0000 (12:56 +0100)]
Add missing includes in header files

This fixes various issues found by globally reordering the include
sections of all .c files.

4 years agotest-util: remove superfluous const
Thomas Hindoe Paaboel Andersen [Thu, 12 Feb 2015 19:32:03 +0000 (20:32 +0100)]
test-util: remove superfluous const

4 years agoremove unused variable
Thomas Hindoe Paaboel Andersen [Thu, 12 Feb 2015 14:12:11 +0000 (15:12 +0100)]
remove unused variable

4 years agoexec: also evaluate working_directory_missing_ok when not applying chroots
Lennart Poettering [Thu, 12 Feb 2015 17:58:43 +0000 (18:58 +0100)]
exec: also evaluate working_directory_missing_ok when not applying chroots

4 years agoNEWS: begin collecting news items for 219
Lennart Poettering [Thu, 12 Feb 2015 17:44:46 +0000 (18:44 +0100)]
NEWS: begin collecting news items for 219

4 years agonetworkd: explicitly enable/disable ipv6ll addresses
Tom Gundersen [Thu, 12 Feb 2015 13:15:41 +0000 (14:15 +0100)]
networkd: explicitly enable/disable ipv6ll addresses

The kernel does not like getting an empty container, so just always pass the parameter, even if
it is the default.

4 years agohwdb: adjust --help text to be more like other --help texts
Lennart Poettering [Thu, 12 Feb 2015 12:17:08 +0000 (13:17 +0100)]
hwdb: adjust --help text to be more like other --help texts

Fix capitalization, indentation, make sure we support --version, and
similar.

4 years agocore: disarm shutdown watchdog if we fail to set timeout
Lennart Poettering [Thu, 12 Feb 2015 11:28:48 +0000 (12:28 +0100)]
core: disarm shutdown watchdog if we fail to set timeout

Better safe than sorry, if drivers are stupid, and reset immediately on
device closing if the timeout could not be initialized.

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=777735

4 years agocore: don't fail to run services in --user instances if $HOME is missing
Lennart Poettering [Thu, 12 Feb 2015 11:21:16 +0000 (12:21 +0100)]
core: don't fail to run services in --user instances if $HOME is missing

Otherwise we cannot even invoke systemd-exit.service anymore, thus not
even exit.

https://bugs.freedesktop.org/show_bug.cgi?id=83100
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=759320

4 years agoRevert "tree-wide: Always use recvmsg with MSG_CMSG_CLOEXEC"
Lennart Poettering [Thu, 12 Feb 2015 10:44:48 +0000 (11:44 +0100)]
Revert "tree-wide: Always use recvmsg with MSG_CMSG_CLOEXEC"

This reverts commit d6d810fbf8071f8510450dbacd1d083f37603656.

It's apparently not OK to pass MSG_CMSG_CLOEXEC to recvmsg() of raw
sockets.

4 years agomissing: add __NR_getrandom for powerpc architecture
Marcel Holtmann [Wed, 11 Feb 2015 21:15:13 +0000 (22:15 +0100)]
missing: add __NR_getrandom for powerpc architecture

4 years agocore: don't watch for socket events when service is in SERVICE_STOP
Lennart Poettering [Wed, 11 Feb 2015 20:25:30 +0000 (21:25 +0100)]
core: don't watch for socket events when service is in SERVICE_STOP

If a service has an ExecStop= program that uses the socket the service was
activated by to tell it to terminate, then that should not be used as
trigger to start the service again.

http://lists.freedesktop.org/archives/systemd-devel/2015-February/028058.html

4 years agosystemctl: don't update the reboot parameter if none is given
Michael Olbrich [Sun, 8 Feb 2015 19:29:50 +0000 (20:29 +0100)]
systemctl: don't update the reboot parameter if none is given

Otherwise systemd-reboot.service will remove the parameter that was set
before. This was broken in commit "b986229efe2cc96157aa14c37bab7843311bbef1
systemctl: bugfix for systemctl reboot command with argument"

4 years agocore: correct a log message
Lennart Poettering [Wed, 11 Feb 2015 17:51:35 +0000 (18:51 +0100)]
core: correct a log message

The log message talks specifically about services, though it actually
applies to any kind of unit.

4 years agotree-wide: whenever we include libgen.h, immediately undefine basename()
Lennart Poettering [Wed, 11 Feb 2015 17:50:38 +0000 (18:50 +0100)]
tree-wide: whenever we include libgen.h, immediately undefine basename()

Also, document in adjacent comments and in CODING_STYLE why we do that.

4 years agocore: remove unneeded <libgen.h> include
Cristian Rodríguez [Tue, 10 Feb 2015 00:41:09 +0000 (21:41 -0300)]
core: remove unneeded <libgen.h> include

execute.c only uses basename (the GNU version in <string.h>)

4 years agoRevert "units: add SecureBits"
Lennart Poettering [Wed, 11 Feb 2015 17:28:06 +0000 (18:28 +0100)]
Revert "units: add SecureBits"

This reverts commit 6a716208b346b742053cfd01e76f76fb27c4ea47.

Apparently this doesn't work.

http://lists.freedesktop.org/archives/systemd-devel/2015-February/028212.html

4 years agoata_id: remove unused header files
Robert Milasan [Thu, 5 Feb 2015 13:19:35 +0000 (14:19 +0100)]
ata_id: remove unused header files

Signed-off-by: Robert Milasan <rmilasan@suse.com>
4 years agounits: add SecureBits
Topi Miettinen [Wed, 11 Feb 2015 16:32:14 +0000 (18:32 +0200)]
units: add SecureBits

No setuid programs are expected to be executed, so add
SecureBits=noroot noroot-locked
to unit files.

4 years agoman: fix typo
Torstein Husebø [Wed, 11 Feb 2015 16:22:55 +0000 (17:22 +0100)]
man: fix typo

4 years agonetwork-internal: fix inverted matching
Tom Gundersen [Wed, 11 Feb 2015 12:10:48 +0000 (13:10 +0100)]
network-internal: fix inverted matching

This fixes the test-suite again.

4 years agonetworkd: fdb - fix const warning
Tom Gundersen [Wed, 11 Feb 2015 12:02:58 +0000 (13:02 +0100)]
networkd: fdb - fix const warning

4 years agonetworkd: don't do exit-on-idle if there is something wrong with the dbus connection
Tom Gundersen [Wed, 11 Feb 2015 11:57:58 +0000 (12:57 +0100)]
networkd: don't do exit-on-idle if there is something wrong with the dbus connection

Just fall back to plain old event loop in this (highly unlikely) case.

In the glorious future when kdbus is upstream we can of course drop all this.

4 years agosysv-generator: Skip init scripts for existing native services
Martin Pitt [Wed, 2 Jul 2014 20:00:00 +0000 (22:00 +0200)]
sysv-generator: Skip init scripts for existing native services

This avoids taking the SysV init script enablement state into account if we
have native units. Otherwise systemctl disable on native unit would not
be respected in the presence of an enabled SysV script.

Also, there's no need to do all the parsing and creation of service files if we
already have a native systemd unit for the processed SysV init script.

4 years agoUpdate TODO
Lennart Poettering [Wed, 11 Feb 2015 11:57:08 +0000 (12:57 +0100)]
Update TODO

4 years agonetworkd DHCPv4 logging endian fix
Paul Martin [Wed, 11 Feb 2015 11:50:28 +0000 (11:50 +0000)]
networkd DHCPv4 logging endian fix

On Tue, Feb 10, 2015 at 08:10:43PM +0100, Lennart Poettering wrote:

> Hmm, I think it would be nicer to use be32toh() here instead, since it
> ensures the macro is (to a limited degree) typesafe.
>
> Any chance you could rework that?

From: Paul Martin <paul.martin@codethink.co.uk>
Date: Wed, 11 Feb 2015 11:47:16 +0000
Subject: [PATCH] networkd dhcpv4 logging endian fix

On a big-endian host, systemd-networkd prints out IPv4 network
addresses byte reversed:

Feb 10 16:43:32 hostname systemd-networkd[151]: eth0 : DHCPv4 address 158.1.24.10/16 via 1.1.24.10

The address obtained is 10.24.1.158/16 and the route is

  10.24.0.0/16 dev eth0  src 10.24.1.187

The macro ADDRESS_FMT_VAL() unpacks a "struct in_addr" in a
little-endian specific manner.

This patch forces the passed address into host order, then unpacks it.

On an x86 later than i486, compiled with -O2, the only extra overhead
is a single bswap instruction.

4 years agoman: boilerplate unification
Zbigniew Jędrzejewski-Szmek [Wed, 11 Feb 2015 04:19:26 +0000 (23:19 -0500)]
man: boilerplate unification

4 years agoman: fix reference to glob manpage
Zbigniew Jędrzejewski-Szmek [Wed, 11 Feb 2015 04:19:16 +0000 (23:19 -0500)]
man: fix reference to glob manpage

4 years agoman: add systemd.generator(7)
Zbigniew Jędrzejewski-Szmek [Wed, 11 Feb 2015 04:17:59 +0000 (23:17 -0500)]
man: add systemd.generator(7)

This is largely based on http://www.freedesktop.org/wiki/Software/systemd/Generators/,
and obsoletes that page. It seems that we do a much better
job of keeping man pages up-to-date compared to wiki pages.
Man pages are also easier to find for users.

https://bugs.freedesktop.org/show_bug.cgi?id=89048

4 years agoTODO
Tom Gundersen [Tue, 10 Feb 2015 21:30:50 +0000 (22:30 +0100)]
TODO

4 years agonet: support globbing and disjunction in Match logic
Tom Gundersen [Tue, 10 Feb 2015 17:30:16 +0000 (18:30 +0100)]
net: support globbing and disjunction in Match logic

Match{Name,OrginalName,Type,Driver,Path} can now take a space-separated glob of matches.

4 years agonetworkd: fdb - refactor a bit
Tom Gundersen [Tue, 10 Feb 2015 19:40:41 +0000 (20:40 +0100)]
networkd: fdb - refactor a bit

Pass around Link objcets rather than FdbEntry objects. The link objects have an up-to-date
ifname we can use for logging. match_name sholud _never_ be used for anything except
matching. Firstly, it may be unset (usually is), and secondly it may not be up-to-date.

4 years agonetworkd: bus - expose MatchMAC
Tom Gundersen [Tue, 10 Feb 2015 16:29:40 +0000 (17:29 +0100)]
networkd: bus - expose MatchMAC

4 years agoman: sd_bus_path_encode - fix order of arguments
Tom Gundersen [Tue, 10 Feb 2015 16:28:34 +0000 (17:28 +0100)]
man: sd_bus_path_encode - fix order of arguments

4 years agonetworkd: network-bus - use sd_bus_path_{en,de}code
Tom Gundersen [Tue, 10 Feb 2015 15:36:58 +0000 (16:36 +0100)]
networkd: network-bus - use sd_bus_path_{en,de}code

4 years agoCODING_STYLE: elaborate on O_CLOEXEC a bit
Lennart Poettering [Tue, 10 Feb 2015 20:25:28 +0000 (21:25 +0100)]
CODING_STYLE: elaborate on O_CLOEXEC a bit

4 years agotree-wide: Always use recvmsg with MSG_CMSG_CLOEXEC
Cristian Rodríguez [Tue, 10 Feb 2015 15:06:24 +0000 (12:06 -0300)]
tree-wide: Always use recvmsg with MSG_CMSG_CLOEXEC

4 years agotest-lldp: fix minor OOM
Lennart Poettering [Tue, 10 Feb 2015 20:17:26 +0000 (21:17 +0100)]
test-lldp: fix minor OOM

4 years agotest-lldp: initialize structs when we define them
Lennart Poettering [Tue, 10 Feb 2015 20:15:47 +0000 (21:15 +0100)]
test-lldp: initialize structs when we define them

4 years agotest-lldp: use strndup() where possible
Lennart Poettering [Tue, 10 Feb 2015 20:15:21 +0000 (21:15 +0100)]
test-lldp: use strndup() where possible

4 years agolibsystemd-network: Avoid potential NULL dereference in test-lldp
Philippe De Swert [Sun, 8 Feb 2015 20:21:51 +0000 (22:21 +0200)]
libsystemd-network: Avoid potential NULL dereference in test-lldp

As a malloc0 could fail, doing a strncpy without checking could
cause issues. Adding an assert should be good enough and in line
with other similar routines in the code.

Found with Coverity Fixes: CID#1261402

4 years agoupdate TODO
Lennart Poettering [Tue, 10 Feb 2015 19:46:19 +0000 (20:46 +0100)]
update TODO

4 years agobus-proxyd: initialize ioctl structure only once
Lennart Poettering [Tue, 10 Feb 2015 19:46:09 +0000 (20:46 +0100)]
bus-proxyd: initialize ioctl structure only once

4 years ago.gitignore: add systemd-pull
Vincent Batts [Tue, 10 Feb 2015 11:31:43 +0000 (12:31 +0100)]
.gitignore: add systemd-pull

4 years agojournald: don't specify inline in local functions
Lennart Poettering [Tue, 10 Feb 2015 11:32:54 +0000 (12:32 +0100)]
journald: don't specify inline in local functions

Leave it to the compiler to figure out whether it shall inline stuff or
not.

Only place where using static inline is OK to use is in in header
files, really.

4 years agoshared: untabify
Lennart Poettering [Tue, 10 Feb 2015 11:32:19 +0000 (12:32 +0100)]
shared: untabify

4 years agologind: tell Coverity that we knowingly ignore mkdir()'s return value
Lennart Poettering [Tue, 10 Feb 2015 11:31:56 +0000 (12:31 +0100)]
logind: tell Coverity that we knowingly ignore mkdir()'s return value

4 years agoupdate TODO
Lennart Poettering [Tue, 10 Feb 2015 11:31:31 +0000 (12:31 +0100)]
update TODO

4 years agokeymap: Add microphone mute keymap for several HP laptops
Hui Wang [Tue, 10 Feb 2015 08:42:22 +0000 (16:42 +0800)]
keymap: Add microphone mute keymap for several HP laptops

On these several HP laptops, the microphone mute hotkey is "Fn+F8" and
the scancode for this hotkey is 0x81, but this scancode was mapped to
fn_esc in the HP generic keymap section. To fix this problem, we add
a machine specific keymap section to add the correct keymap rule.

BugLink: https://bugs.launchpad.net/bugs/1409721
BugLink: https://bugs.launchpad.net/bugs/1334968
Signed-off-by: Hui Wang <hui.wang@canonical.com>
4 years agonetworkd: link_object_find - don't accept invalid input
Tom Gundersen [Mon, 9 Feb 2015 21:02:32 +0000 (22:02 +0100)]
networkd: link_object_find - don't accept invalid input

Reported by Zbigniew.

4 years agonetworkd: support route scopes
Tom Gundersen [Mon, 9 Feb 2015 15:22:34 +0000 (16:22 +0100)]
networkd: support route scopes

For now we only support the hardcoded values RT_SCOPE_{UNIVERSE,LOCAL,HOST},
and not numerical values or values from /etc/iproute2/rt_scopes.

This addresses https://bugs.freedesktop.org/show_bug.cgi?id=88508.

4 years agotreewide: correct typos and use consistent "MAC" spelling
Torstein Husebø [Mon, 9 Feb 2015 13:18:57 +0000 (14:18 +0100)]
treewide: correct typos and use consistent "MAC" spelling

4 years agonetworkd: bus - switch to properly escaped object paths
Tom Gundersen [Mon, 9 Feb 2015 11:00:33 +0000 (12:00 +0100)]
networkd: bus - switch to properly escaped object paths

4 years agonetworkd: add support for IPv6 tokens
Tom Gundersen [Sun, 8 Feb 2015 22:20:56 +0000 (23:20 +0100)]
networkd: add support for IPv6 tokens

This allows the admin to set the host-specific part of IPv6 addresses, but still
receive the prefix via SLAAC.

.network file snippet:

[Network]
IPv6Token=::12

gives:

$ ip token
token ::12 dev eth0

This closes https://bugs.freedesktop.org/show_bug.cgi?id=81177.

4 years agonetworkd: generalize IPv4LL to LinkLocal
Tom Gundersen [Sun, 8 Feb 2015 21:27:15 +0000 (22:27 +0100)]
networkd: generalize IPv4LL to LinkLocal

This allows both IPv4 and IPv6 link-local addresses to be enabled or disabled. By default
we still enable IPv6LL and disable IPv4LL. The old config option is kept for backwards
compatibility, but removed from the documentation.

4 years agosd-rtnl: fix typesystem for IFLA_AF_SPEC
Tom Gundersen [Sun, 8 Feb 2015 21:34:53 +0000 (22:34 +0100)]
sd-rtnl: fix typesystem for IFLA_AF_SPEC

Got this one wrong, it is not a union, just a nested container.

4 years agolanguage fallback: it_CH (Italian, Swiss) -> it_IT (Italian, Italy)
Daniele Medri [Sun, 8 Feb 2015 07:57:49 +0000 (08:57 +0100)]
language fallback: it_CH (Italian, Swiss) -> it_IT (Italian, Italy)

4 years agonetworkd: add basic org.freedesktop.network1.Network interface
Tom Gundersen [Sun, 8 Feb 2015 12:27:56 +0000 (13:27 +0100)]
networkd: add basic org.freedesktop.network1.Network interface

4 years agonetworkd: add network_get_by_name
Tom Gundersen [Sun, 8 Feb 2015 12:29:35 +0000 (13:29 +0100)]
networkd: add network_get_by_name

4 years agobuild-sys: properly enable the networkd dbus activation
Tom Gundersen [Sun, 8 Feb 2015 18:43:01 +0000 (19:43 +0100)]
build-sys: properly enable the networkd dbus activation

4 years agosd-rtnl: add support for IFLA_INET6_*
Tom Gundersen [Sun, 8 Feb 2015 18:31:46 +0000 (19:31 +0100)]
sd-rtnl: add support for IFLA_INET6_*

4 years agosd-rtnl: extend type system to allow address-family to decide the union members
Tom Gundersen [Sun, 8 Feb 2015 11:37:05 +0000 (12:37 +0100)]
sd-rtnl: extend type system to allow address-family to decide the union members

So far we only supported selecting them by sibling attributes.

(This stuff is all a bit crazy, but there seems to be no other way...)

4 years agonetworkd: use valid bus paths
Mantas Mikulėnas [Fri, 6 Feb 2015 20:35:14 +0000 (22:35 +0200)]
networkd: use valid bus paths

Object path components must start with [A-Za-z_] (AFAIK).
Also the value of 'p' is undefined if asprintf fails.
Compare to user_bus_path() in src/login/logind-user-dbus.c:281.

4 years agosd-rtnl: don't treat 0-length messages specially
Tom Gundersen [Fri, 6 Feb 2015 13:43:17 +0000 (14:43 +0100)]
sd-rtnl: don't treat 0-length messages specially

Still parse the CMSG data, and most importantly make sure we drop the message when peeking.

4 years agosysctl: consider --prefix while parsing the files
Umut Tezduyar Lindskog [Sat, 7 Feb 2015 13:12:41 +0000 (14:12 +0100)]
sysctl: consider --prefix while parsing the files

not while applying the parsed sysctl values. Otherwise
info "Overwriting earlier assignment of %s in file %s" is
visible many times even though the given --prefix doesn't
try to set the overridden value.

This also optimizes the startup tiny bit since we have udev
rules running on network devices and setting sysctl through
the rules.

4 years agoAdd Silesian and Kashubian to language fallback map
Piotr Drąg [Sat, 7 Feb 2015 16:33:40 +0000 (17:33 +0100)]
Add Silesian and Kashubian to language fallback map

Silesian and Kashubian speakers would prefer to fall back to Polish (pl)
translations instead of English (C).

https://bugs.freedesktop.org/show_bug.cgi?id=89021

4 years agobus-proxyd: fix 'ListQueuedOwners' call
Lukasz Skalski [Fri, 6 Feb 2015 16:08:32 +0000 (17:08 +0100)]
bus-proxyd: fix 'ListQueuedOwners' call

Set proper kdbus_cmd_list object size, otherwise:

dbus-send --system --dest=org.freedesktop.DBus --type=method_call \
print-reply / org.freedesktop.DBus.ListQueuedOwners string:org.freedesktop.systemd1

Error org.freedesktop.DBus.Error.InvalidArgs: Invalid argument

4 years agolocaled: add LANGUAGE= fallback when LANG= is specified
Zbigniew Jędrzejewski-Szmek [Sat, 31 Jan 2015 19:11:23 +0000 (14:11 -0500)]
localed: add LANGUAGE= fallback when LANG= is specified

For the entries listed in the first column of language-fallback-map,
the entry from the second column will be used for LANGUAGE=, if
LANGUAGE= is not explicitly specified.

https://bugzilla.redhat.com/show_bug.cgi?id=624158

4 years agoAdd language fallback map
Naveen Kumar [Sat, 31 Jan 2015 17:50:49 +0000 (12:50 -0500)]
Add language fallback map

This map will be used to provide a fallback for translations.
For example, a Niederdeutsch (nds) speaker prefers to fall back to
German (de) translations rather then the English (C) ones.

https://bugzilla.redhat.com/show_bug.cgi?id=624158#c9

4 years agobuild-sys: remove reference to long gone file
Zbigniew Jędrzejewski-Szmek [Fri, 6 Feb 2015 14:12:27 +0000 (09:12 -0500)]
build-sys: remove reference to long gone file

Interesting that automake doesn't complain about this at all.
That file was removed in 3fb97a58fa3f233.

4 years agounits: networkd - fix busname to work on kdbus
Tom Gundersen [Fri, 6 Feb 2015 11:09:36 +0000 (12:09 +0100)]
units: networkd - fix busname to work on kdbus

4 years agosd-rtnl: don't fall over when receiving 0 bytes from the socket
Tom Gundersen [Fri, 6 Feb 2015 10:54:30 +0000 (11:54 +0100)]
sd-rtnl: don't fall over when receiving 0 bytes from the socket

causes EOF. Seems like a kernel bug. Ignoring it seems to work be the best we
can do for now...

See https://bugs.freedesktop.org/show_bug.cgi?id=88397

4 years agosd-rtnl: fix creation of synthetic error replies
Tom Gundersen [Fri, 6 Feb 2015 10:49:07 +0000 (11:49 +0100)]
sd-rtnl: fix creation of synthetic error replies

4 years agoudev: net_setup - clarify reason for failure of persistent mac address policy
Tom Gundersen [Fri, 6 Feb 2015 07:36:17 +0000 (08:36 +0100)]
udev: net_setup - clarify reason for failure of persistent mac address policy

4 years agoshared/log: read /proc/cmdline only in daemons
Zbigniew Jędrzejewski-Szmek [Fri, 6 Feb 2015 00:44:13 +0000 (19:44 -0500)]
shared/log: read /proc/cmdline only in daemons

http://lists.freedesktop.org/archives/systemd-devel/2015-February/027960.html

4 years agoupdate TODO
Lukas Nykryn [Thu, 5 Feb 2015 17:07:54 +0000 (18:07 +0100)]
update TODO

4 years agonetworkd: move the connection to the bus out of manager_new (again)
Tom Gundersen [Thu, 5 Feb 2015 17:00:16 +0000 (18:00 +0100)]
networkd: move the connection to the bus out of manager_new (again)

This would otherwise make the tests fail as we cannot grab the bus name.

4 years agonetworkd: gracefully handle failure to emit signals when dbus is not connected
Tom Gundersen [Thu, 5 Feb 2015 16:59:40 +0000 (17:59 +0100)]
networkd: gracefully handle failure to emit signals when dbus is not connected

When we get kdbus we don't need these hackse.

4 years agobus: sync with kdbus (ABI break)
David Herrmann [Wed, 4 Feb 2015 15:21:27 +0000 (16:21 +0100)]
bus: sync with kdbus (ABI break)

4 years agonetworkd: exit on idle
Tom Gundersen [Wed, 4 Feb 2015 14:00:20 +0000 (15:00 +0100)]
networkd: exit on idle

We will be woken up on rtnl or dbus activity, so let's just quit if some time has passed and that is the only thing that can happen.

Note that we will always stay around if we expect network activity (e.g. DHCP is enabled), as we are not restarted on that.

4 years agonetworkd: don't unlink state files on exit
Tom Gundersen [Wed, 4 Feb 2015 14:16:40 +0000 (15:16 +0100)]
networkd: don't unlink state files on exit

Only unlink state files on DELLINK. This allows sd-network to be used even when networkd is not running.

4 years agonetworkd: add basic dbus API
Tom Gundersen [Wed, 4 Feb 2015 10:44:37 +0000 (11:44 +0100)]
networkd: add basic dbus API

Only the very basics, more to come.

For now:

$ busctl tree org.freedesktop.network1
└─/org/freedesktop/network1
  └─/org/freedesktop/network1/link
    ├─/org/freedesktop/network1/link/1
    ├─/org/freedesktop/network1/link/2
    ├─/org/freedesktop/network1/link/3
    ├─/org/freedesktop/network1/link/4
    ├─/org/freedesktop/network1/link/5
    ├─/org/freedesktop/network1/link/6
    ├─/org/freedesktop/network1/link/7
    ├─/org/freedesktop/network1/link/8
    └─/org/freedesktop/network1/link/9

$ busctl introspect org.freedesktop.network1 /org/freedesktop/network1
NAME                                TYPE      SIGNATURE RESULT/VALUE FLAGS
org.freedesktop.network1.Manager    interface -         -            -
.OperationalState                   property  s         "carrier" emits-change

$ busctl introspect org.freedesktop.network1 /org/freedesktop/network1/link/1
NAME                                TYPE      SIGNATURE RESULT/VALUE FLAGS
org.freedesktop.network1.Link       interface -         -            -
.AdministrativeState                property  s         "unmanaged" emits-change
.OperationalState                   property  s         "carrier" emits-change

4 years agonetworkd: don't warn about missing links unnecessarily
Tom Gundersen [Wed, 4 Feb 2015 09:08:12 +0000 (10:08 +0100)]
networkd: don't warn about missing links unnecessarily

If we get a NEWLINK + NEWADDR between enumerating the links and enumerating the addresses, we
would get a warning that the link corresponding to the address does not exist. This is a false
warning as both the NEWLINK and NEWADDR would be processed after enumerating completed, so drop
it.

4 years agonetworkd: log when finished enumerating links and addresses
Tom Gundersen [Wed, 4 Feb 2015 08:47:50 +0000 (09:47 +0100)]
networkd: log when finished enumerating links and addresses

4 years agosystemctl: unit_find_paths(): unify error handling in two code pathes
Ivan Shapovalov [Wed, 4 Feb 2015 22:57:00 +0000 (01:57 +0300)]
systemctl: unit_find_paths(): unify error handling in two code pathes

4 years agosystemctl: cat, edit: further polish error messages
Ivan Shapovalov [Wed, 4 Feb 2015 22:56:59 +0000 (01:56 +0300)]
systemctl: cat, edit: further polish error messages

4 years agosystemctl: cat: fix error handling
Ivan Shapovalov [Wed, 4 Feb 2015 22:56:58 +0000 (01:56 +0300)]
systemctl: cat: fix error handling

- correctly check for local vs. remote transport
- return after receiving error from expand_names()

4 years agosystemctl: cat, edit: improve unit load error reporting
Ivan Shapovalov [Wed, 4 Feb 2015 22:56:57 +0000 (01:56 +0300)]
systemctl: cat, edit: improve unit load error reporting

- report actual load error for units which could not be loaded
- make unit_find_paths() report all kinds of errors it encounters
  (for consistency)
- consistently handle not-found errors in cat() and edit()

4 years agohwdb: Bind "toolbox" buttons to the Windows key
Bastien Nocera [Wed, 4 Feb 2015 17:46:10 +0000 (18:46 +0100)]
hwdb: Bind "toolbox" buttons to the Windows key

One would expect pressing the button to go to an overview / show
applications mode, we thus map it to leftmeta, the Windows key.

See https://bugzilla.gnome.org/show_bug.cgi?id=658602#c17

4 years agobuild-sys: mark systemd-firstboot man page as conditional
Michael Biebl [Wed, 4 Feb 2015 20:31:26 +0000 (21:31 +0100)]
build-sys: mark systemd-firstboot man page as conditional

Rebuild Makefile-man.am accordingly.

4 years agotimesyncd: downgrade more log messages from LOG_INFO to LOG_DEBUG
Lennart Poettering [Wed, 4 Feb 2015 16:00:23 +0000 (17:00 +0100)]
timesyncd: downgrade more log messages from LOG_INFO to LOG_DEBUG

https://bugs.freedesktop.org/show_bug.cgi?id=87505

Let's make timesyncd less chatty.

4 years agoshared/capabilities: simplify assertions about bits
Zbigniew Jędrzejewski-Szmek [Wed, 4 Feb 2015 15:24:02 +0000 (10:24 -0500)]
shared/capabilities: simplify assertions about bits

The assert added in 7d328b5446 was wrong. Also update the comments
and make sure we don't try to shift by type size.

4 years agoshared/util: drop duplicate log2u64 function
Zbigniew Jędrzejewski-Szmek [Wed, 4 Feb 2015 14:39:36 +0000 (09:39 -0500)]
shared/util: drop duplicate log2u64 function

4 years agoman: fix diagrams in bootup(7)
Zbigniew Jędrzejewski-Szmek [Wed, 4 Feb 2015 14:35:37 +0000 (09:35 -0500)]
man: fix diagrams in bootup(7)

Fallout from 798d3a524e.

4 years agotimesyncd: reduce log chattiness a bit
Lennart Poettering [Wed, 4 Feb 2015 13:46:40 +0000 (14:46 +0100)]
timesyncd: reduce log chattiness a bit

Let's downgrade the statistics output from LOG_INFO to LOG_DEBUG.

https://bugs.freedesktop.org/show_bug.cgi?id=88926

4 years agonspawn: Allow module loading if CAP_SYS_MODULE is requested
Jay Faulkner [Wed, 4 Feb 2015 01:45:50 +0000 (17:45 -0800)]
nspawn: Allow module loading if CAP_SYS_MODULE is requested

nspawn containers currently block module loading in all cases, with
no option to disable it. This allows an admin, specifically setting
capability=CAP_SYS_MODULE or capability=all to load modules.

4 years agoshared/capability: don't be too frugal on space for caps
Tom Gundersen [Wed, 4 Feb 2015 08:23:24 +0000 (09:23 +0100)]
shared/capability: don't be too frugal on space for caps

We were dropping the most significant bit. Add an assert to make sure it does not happen again.

Fixes a bug introduced in 7d328b544621d4b1bec936dec612947ad8bfb65a.