Lennart Poettering [Thu, 27 Aug 2015 20:30:43 +0000 (22:30 +0200)]
core: add unit_dbus_interface_from_type() to unit-name.h
Let's add a way to get the type-specific D-Bus interface of a unit from
either its type or name to src/basic/unit-name.[ch]. That way we can
share it with the client side, where it is useful in tools like cgls or
machinectl.
Also ports over machinectl to make use of this.
Lennart Poettering [Thu, 27 Aug 2015 19:05:13 +0000 (21:05 +0200)]
update TODO
David Herrmann [Thu, 27 Aug 2015 15:03:18 +0000 (17:03 +0200)]
NEWS: update
Add 'machinectl shell' to NEWS and also update date and contributors
lists.
David Herrmann [Thu, 27 Aug 2015 14:43:02 +0000 (16:43 +0200)]
sd-bus: do not connect to dbus-1 socket when kdbus is available
We should not fall back to dbus-1 and connect to the proxy when kdbus
returns an error that indicates that kdbus is running but just does not
accept new connections because of quota limits or something similar.
Based on a patch by Kay.
David Herrmann [Thu, 27 Aug 2015 14:32:22 +0000 (16:32 +0200)]
Revert "sd-bus: do not connect to dbus-1 socket when kdbus is available"
This reverts commit
d4d00020d6ad855d65d31020fefa5003e1bb477f. The idea of
the commit is broken and needs to be reworked. We really cannot reduce
the bus-addresses to a single address. We always will have systemd with
native clients and legacy clients at the same time, so we also need both
addresses at the same time.
David Herrmann [Thu, 27 Aug 2015 14:23:47 +0000 (16:23 +0200)]
selinux: drop mac_selinux_unit_access_check_strv()
It is not acceptable to load unit files during enable/disable operations
just to figure out the selinux labels. systemd implements lazy loading
for units, so the selinux hooks need to follow it.
This drops the mac_selinux_unit_access_check_strv() helper which
implements a non-acceptable policy check. If anyone cares for that
functionality, you really should pass a callback+userdata to the helpers
in src/shared/install.c which does policy checks on each touched file.
See #1050 on github for more.
David Herrmann [Thu, 27 Aug 2015 14:22:33 +0000 (16:22 +0200)]
sd-bus: rename bloom-tag to arg0-has
We use dashes in our bloom-tags. Make sure the newly introduced arg0has
tag uses the same style.
Note that the external dbus-tags don't use dashes, though. They are
defined in the spec and we need to keep compatibility there.
David Herrmann [Thu, 27 Aug 2015 14:14:20 +0000 (16:14 +0200)]
Merge pull request #1036 from poettering/sd-bus-arg0has
sd-bus: introduce new match type "arg0has=" for matching arrays of st…
David Herrmann [Thu, 27 Aug 2015 14:12:52 +0000 (16:12 +0200)]
Merge pull request #1024 from poettering/sd-bus-explicit
sd-bus: introduce new SD_BUS_VTABLE_PROPERTY_EXPLICIT flag
Kay Sievers [Thu, 27 Aug 2015 10:57:17 +0000 (12:57 +0200)]
udev: control - add FIXME comment to remove work-around
Kay Sievers [Wed, 26 Aug 2015 16:37:16 +0000 (18:37 +0200)]
hwdb: update
Kay Sievers [Wed, 26 Aug 2015 14:09:17 +0000 (16:09 +0200)]
build-sys: bump versions
David Herrmann [Wed, 26 Aug 2015 10:37:56 +0000 (12:37 +0200)]
sd-dhcp6: fix domainname memleak
strv_extend() does not consume the passed entry, hence, we must properly
free it. Furthermore, we should *not* use strv_consume() as we do greedy
allocations on 'ret'; and greedy-allocations should only be used for short
lived objects or caches.
Fix the domainname parser to properly free temporary storage when done.
David Herrmann [Wed, 26 Aug 2015 10:30:56 +0000 (12:30 +0200)]
sd-dhcp: don't randomly ref objects
In our API design, getter-functions don't ref objects. Calls like
foo_get_bar() will not ref 'bar'. We never do that and there is no real
reason to do it in single threaded APIs. If you need a ref-count, you
better take it yourself *BEFORE* doing anything else on the parent object
(as this might invalidate your pointer).
Right now, sd_dhcp?_get_lease() refs the lease it returns. A lot of
code-paths in systemd do not expect this and thus leak the lease
reference. Fix this by changing the API to not ref returned objects.
Daniel Mack [Wed, 26 Aug 2015 10:01:56 +0000 (12:01 +0200)]
Merge pull request #1048 from poettering/resolved-man
man: resolved man page updates (plus one to machinectl)
David Herrmann [Wed, 26 Aug 2015 09:55:48 +0000 (11:55 +0200)]
Merge pull request #1047 from poettering/machinectl-TERM
machinectl: pass $TERM into "machinectl shell" sessions
Lennart Poettering [Wed, 26 Aug 2015 09:02:28 +0000 (11:02 +0200)]
man: minor extension to the machinectl man page
s/an/any/, as reported by Vito Caputo.
Also mention explicitly that the security properties (i.e. SELinux) are
also isolated when "machinectl shell" is used.
Lennart Poettering [Wed, 26 Aug 2015 09:00:09 +0000 (11:00 +0200)]
man: document nss-resolve
Martin Pitt [Wed, 26 Aug 2015 08:53:45 +0000 (10:53 +0200)]
Merge pull request #1044 from d-hatayama/fix_systemctl_abspath
selinux: fix regression of systemctl subcommands when absolute unit file paths are specified
Lennart Poettering [Wed, 26 Aug 2015 08:30:06 +0000 (10:30 +0200)]
man: document resolved's RR synthesizing and query routing
Lennart Poettering [Wed, 26 Aug 2015 08:09:11 +0000 (10:09 +0200)]
Merge pull request #998 from vbatts/tar_nosparse_flag
import: don't create sparse tar archives
Lennart Poettering [Wed, 26 Aug 2015 08:07:21 +0000 (10:07 +0200)]
machinectl: pass $TERM into "machinectl shell" sessions
Tom Gundersen [Wed, 26 Aug 2015 08:02:37 +0000 (10:02 +0200)]
Merge pull request #1046 from poettering/resolved-dump
dump resolved cache/zone info on SIGSUR1, plus one fix
Lennart Poettering [Wed, 26 Aug 2015 07:42:06 +0000 (09:42 +0200)]
manager: close hostname fd *after* removing it epoll
Otherwise the epoll removal will fail and result in a warning.
Lennart Poettering [Wed, 26 Aug 2015 07:41:45 +0000 (09:41 +0200)]
resolved: dump cache and zone contents to syslog on SIGUSR1
Marcel Holtmann [Wed, 26 Aug 2015 07:21:02 +0000 (09:21 +0200)]
hwdb: Update database of Bluetooth company identifiers
HATAYAMA Daisuke [Wed, 26 Aug 2015 03:07:31 +0000 (12:07 +0900)]
selinux: fix regression of systemctl subcommands when absolute unit file paths are specified
The commit
4938696301a914ec26bcfc60bb99a1e9624e3789 overlooked the
fact that unit files can be specified as unit file paths, not unit
file names, wrongly passing a unit file path to the 1st argument of
manager_load_unit() that handles it as a unit file name. As a result,
the following 4 systemctl subcommands:
enable
disable
reenable
link
mask
unmask
fail with the following error message:
# systemctl enable /usr/lib/systemd/system/kdump.service
Failed to execute operation: Unit name /usr/lib/systemd/system/kdump.service is not valid.
# systemctl disable /usr/lib/systemd/system/kdump.service
Failed to execute operation: Unit name /usr/lib/systemd/system/kdump.service is not valid.
# systemctl reenable /usr/lib/systemd/system/kdump.service
Failed to execute operation: Unit name /usr/lib/systemd/system/kdump.service is not valid.
# cp /usr/lib/systemd/system/kdump.service /tmp/
# systemctl link /tmp/kdump.service
Failed to execute operation: Unit name /tmp/kdump.service is not valid.
# systemctl mask /usr/lib/systemd/system/kdump.service
Failed to execute operation: Unit name /usr/lib/systemd/system/kdump.service is not valid.
# systemctl unmask /usr/lib/systemd/system/kdump.service
Failed to execute operation: Unit name /usr/lib/systemd/system/kdump.service is not valid.
To fix the issue, first check whether a unit file is passed as a unit
file name or a unit file path, and then pass the unit file to the
appropreate argument of manager_load_unit().
By the way, even with this commit mask and unmask reject unit file
paths as follows and this is a correct behavior:
# systemctl mask /usr/lib/systemd/system/kdump.service
Failed to execute operation: Invalid argument
# systemctl unmask /usr/lib/systemd/system/kdump.service
Failed to execute operation: Invalid argument
Daniel Mack [Tue, 25 Aug 2015 22:05:33 +0000 (00:05 +0200)]
Merge pull request #1043 from phomes/master
test-util: fix a memleak
Daniel Mack [Tue, 25 Aug 2015 21:58:40 +0000 (23:58 +0200)]
Merge pull request #1039 from poettering/nspawn-machine-template
nspawn: make sure --template= and --machine= my be combined
Daniel Mack [Tue, 25 Aug 2015 21:58:03 +0000 (23:58 +0200)]
Merge pull request #1038 from poettering/coredumpctl-directory
Add --directory= option for reading alternate journal
Daniel Mack [Tue, 25 Aug 2015 21:57:28 +0000 (23:57 +0200)]
Merge pull request #1040 from poettering/cgroup-path-fix
fix "systemctl status idontexist.service" showing the full cgroup tree
Thomas Hindoe Paaboel Andersen [Tue, 25 Aug 2015 21:50:02 +0000 (23:50 +0200)]
test-util: fix a memleak
Daniel Mack [Tue, 25 Aug 2015 21:51:34 +0000 (23:51 +0200)]
resolve: fix regression in dns-scope
Bring back a return statement
106784eb errornously removed.
Thanks to @phomes for reporting.
Daniel Mack [Tue, 25 Aug 2015 21:48:07 +0000 (23:48 +0200)]
Merge pull request #1041 from phomes/master
trivial cleanups
Daniel Mack [Tue, 25 Aug 2015 21:47:30 +0000 (23:47 +0200)]
Merge pull request #1034 from poettering/resolved-fixes-2
various resolved fixes
Thomas Hindoe Paaboel Andersen [Tue, 25 Aug 2015 19:10:21 +0000 (21:10 +0200)]
machinectl: remove unused variable
Thomas Hindoe Paaboel Andersen [Tue, 25 Aug 2015 19:07:41 +0000 (21:07 +0200)]
execute: make the invalid entry of the enum -1
Set _EXEC_UTMP_MODE_INVALID to -1. This matches the return value from
string_table_lookup.
Lennart Poettering [Tue, 25 Aug 2015 18:42:50 +0000 (20:42 +0200)]
core: report root cgroup as "/" over the bus
Internally, the root cgroup is stored as the empty string in
Unit.cgroup_path, and "no cgroup" as NULL. Unfortunately, D-Bus does not
know a NULL concept, hence when reporting the cgroup to clients we
should turn the root cgroup into "/", and leave the empty string for the
"no cgroup" case.
This should make sure that "systemctl status -- -.slice" works correctly
and shows the entire cgroup tree.
Lennart Poettering [Tue, 25 Aug 2015 18:37:54 +0000 (20:37 +0200)]
Revert "systemctl: properly handle empty control group paths in "status""
This reverts commit
b04c25f9ef6359ed0ae403bdbfe4df840aba0f58.
Lennart Poettering [Tue, 25 Aug 2015 18:26:51 +0000 (20:26 +0200)]
nspawn: make sure --template= and --machine= my be combined
Fixes #1018.
Based on a patch from Seth Jennings.
Lennart Poettering [Tue, 25 Aug 2015 18:04:55 +0000 (20:04 +0200)]
man: minor fixes to coredumpctl(1)
Stef Walter [Tue, 25 Aug 2015 08:36:49 +0000 (10:36 +0200)]
coredumpctl: Add --directory option for reading alternate journal
In the Cockpit integration tests we hang onton the journal files
for a failed test and would like to inspect them using coredumpctl.
This commit adds the ability to specify an alternate directory
for coredumpctl to read the journal from.
Lennart Poettering [Tue, 25 Aug 2015 17:28:30 +0000 (19:28 +0200)]
sd-bus: introduce new match type "arg0has=" for matching arrays of strings
Previously, sd-bus inofficially already supported bus matches that
tested a string against an array of strings ("as"). This was done via an
enhanced way to interpret "arg0=" matches. This is problematic however,
since clients have no way to determine if their respective
implementation understood strv matches or not, thus allowing invalid
matches to be installed without a way to detect that.
This patch changes the logic to only allow such matches with a new
"arg0has=" syntax. This has the benefit that non-conforming
implementations will return a parse error and a client application may
thus efficiently detect support for the match type.
Matches of this type are useful for "udev"-like systems that "tag" objects
with a number of strings, and clients need to be able to match against
any of these "tags".
The name "has" takes inspiration from Python's ".has_key()" construct.
Lennart Poettering [Tue, 25 Aug 2015 16:59:40 +0000 (18:59 +0200)]
resolved: add comments to DNS_PACKET_MAKE_FLAGS() clarifying DNS vs LLMNR
Some flags are defined differently on unicast DNS and LLMNR, let's
document this in the DNS_PACKET_MAKE_FLAGS() macro.
Lennart Poettering [Tue, 25 Aug 2015 16:54:55 +0000 (18:54 +0200)]
resolved: make packet flags logic more expressive again
This partially reverts
106784ebb7b303ae471851100a773ad2aebf5b80, ad
readds separate DNS_PACKET_MAKE_FLAGS() invocations for the LLMNR and
DNS case. This is important since SOme flags have different names and
meanings on LLMNR and on DNS and we should clarify that via the comments
and how we put things together.
Lennart Poettering [Tue, 25 Aug 2015 16:51:21 +0000 (18:51 +0200)]
resolved: rename DNS UDP socket to 'dns_udp_fd'
This hopefully makes this a bit more expressive and clarifies that the
fd is not used for the DNS TCP socket. This also mimics how the LLMNR
UDP fd is named in the manager object.
Tom Gundersen [Tue, 25 Aug 2015 16:50:44 +0000 (18:50 +0200)]
Merge pull request #1033 from poettering/machine-fixes
misc machine fixes
Lennart Poettering [Tue, 25 Aug 2015 16:39:50 +0000 (18:39 +0200)]
bus-util: make more properties settable in --property=
Add a couple of new properties to the supported set we can pass in
systemd-run's and systemd-nspawn's --property= switch.
Lennart Poettering [Tue, 25 Aug 2015 16:37:53 +0000 (18:37 +0200)]
man: add hyphen to improve man text
Lennart Poettering [Tue, 25 Aug 2015 16:37:31 +0000 (18:37 +0200)]
core: drop spurious new line
Lennart Poettering [Tue, 25 Aug 2015 16:37:16 +0000 (18:37 +0200)]
machine: policykit string fixes
Tom Gundersen [Tue, 25 Aug 2015 14:46:27 +0000 (16:46 +0200)]
Merge pull request #1022 from poettering/machinectl-shell
Add new "machinectl shell" command for su(1)-like behaviour
Lennart Poettering [Tue, 25 Aug 2015 13:54:23 +0000 (15:54 +0200)]
Merge pull request #1029 from jsynacek/unprivileged-wall-message-v2
logind/systemctl: introduce SetWallMessage and --message
Daniel Mack [Tue, 25 Aug 2015 13:01:51 +0000 (15:01 +0200)]
NEWS: preparations for v225 release
Tom Gundersen [Tue, 25 Aug 2015 12:33:41 +0000 (14:33 +0200)]
Merge pull request #1030 from zonque/resolve
resolve: minor cleanups
Daniel Mack [Tue, 4 Aug 2015 11:53:02 +0000 (13:53 +0200)]
resolved: allow dns_cache_put() without a question
Currently, dns_cache_put() does a number of things:
1) It unconditionally removes all keys contained in the passed
question before adding keys from the newly arrived answers.
2) It puts positive entries into the cache for all RRs contained
in the answer.
3) It creates negative entries in the cache for all keys in the
question that are not answered.
Allow passing q = NULL in the parameters and skip 1) and 3), so
we can use that function for mDNS responses. In this case, the
question is irrelevant, we are interested in all answers we got.
Daniel Mack [Tue, 28 Jul 2015 13:00:59 +0000 (15:00 +0200)]
sd-network: make LLMNR specific config parser generic
Rename the enum, the lookup functions and the parser for LLMNRSupport so
the type can be reused for mDNS.
Daniel Mack [Tue, 4 Aug 2015 08:37:59 +0000 (10:37 +0200)]
resolved: move assertion
Make a scope with invalid protocol state fail as soon as possible.
Daniel Mack [Sat, 11 Jul 2015 20:21:26 +0000 (16:21 -0400)]
resolved: use switch-case statements for protocol details
With more protocols to come, switch repetitive if-else blocks with a
switch-case statements.
Jan Synacek [Mon, 24 Aug 2015 12:54:22 +0000 (14:54 +0200)]
logind/systemctl: introduce SetWallMessage and --message
Enable unprivileged users to set wall message on a shutdown
operation. When the message is set via the --message option,
it is logged together with the default shutdown message.
$ systemctl reboot --message "Applied kernel updates."
$ journalctl -b -1
...
systemd-logind[27]: System is rebooting. (Applied kernel updates.)
...
Daniel Mack [Tue, 28 Jul 2015 16:09:08 +0000 (18:09 +0200)]
resolved: remove runtime check for previously asserted condition
Tom Gundersen [Tue, 25 Aug 2015 08:18:16 +0000 (10:18 +0200)]
Merge pull request #1023 from poettering/resolved-fixes
A variety of resolved fixes
Lennart Poettering [Mon, 24 Aug 2015 23:45:33 +0000 (01:45 +0200)]
sd-bus: introduce new SD_BUS_VTABLE_PROPERTY_EXPLICIT flag
This allows marking properties as "explicit". Properties marked like
this are included in the introspection, but are avoided in GetAll()
property queries, PropertiesChanged() signals and in in GetManaged()
object manager calls and InterfacesAdded() signals.
Expensive properties may be marked that way, and they will be
retrievable when explicitly being requested, but never in "blanket"
all-property queries and signals.
This flag may be combined with the flags for "const" and
"emit-validation" properties, but not with "emit-validation", as that
is only useful for properties whose value shall be sent in "blanket"
all-property signals.
The "explicit" flag is also exposed in the introspection data via a new
annotation.
Lennart Poettering [Mon, 24 Aug 2015 21:47:28 +0000 (23:47 +0200)]
resolved: change error code when trying to resolve direct LLMNR PTR RRs
If we try to resoolve an LLMNR PTR RR we shall connect via TCP directly
to the specified IP address. We already refuse to do this if the address
to resolve is of a different address family as the transaction's scope.
The error returned was EAFNOSUPPORT. Let's change this to ESRCH which is
how we indicate "not server available" when connecting for LLMNR or DNS,
since that's what this really is: we have no server we could connect to
in this address family.
This allows us to ensure that no server errors are always handled the same
way.
Lennart Poettering [Mon, 24 Aug 2015 21:46:24 +0000 (23:46 +0200)]
resolve-host: support parsing numeric interface names
If the user specifies an interface by its ifindex we should handle this
nicely. Hence let's try to parse the ifindex as a number before we try
to resolve it as an interface name.
Lennart Poettering [Mon, 24 Aug 2015 21:44:33 +0000 (23:44 +0200)]
resolved: remove duplicate handling of "no servers" query result
So far we handled immediate "no server" query results differently from
"no server" results we ran into during operation: the former would cause
the dns_query_go() call to fail with ESRCH, the later would result in
the query completion callback to be called.
Remove the duplicate codepaths, by always going through the completion
callback. This allows us to remove quite a number of lines for handling
the ESRCH.
This commit should not alter behaviour at all.
Lennart Poettering [Mon, 24 Aug 2015 21:15:51 +0000 (23:15 +0200)]
resolved: replace transaction list by hashmap
Right now we keep track of ongoing transactions in a linked listed for
each scope. Replace this by a hashmap that is indexed by the RR key.
Given that all ongoing transactions will be placed in pretty much the
same scopes usually this should optimize behaviour.
We used to require a list here, since we wanted to do "superset" query
checks, but this became obsolete since transactions are now single-key
instead of multi-key.
Lennart Poettering [Mon, 24 Aug 2015 20:44:54 +0000 (22:44 +0200)]
machinectl: extend the "shell" syntax to take user@container names
In order to make "machinectl shell" more similar to ssh, allow the
following syntax to connect to a container under a specific username:
machinectl shell lennart@fedora
Also beefs up related man page documentation.
Lennart Poettering [Mon, 24 Aug 2015 20:17:52 +0000 (22:17 +0200)]
machinectl: make machine name parameters for "shell" and "login" optional
If no machine name is specified, imply that we connect to ".host", i.e.
the local host.
Lennart Poettering [Mon, 24 Aug 2015 19:27:37 +0000 (21:27 +0200)]
machined: beef up PolicyKit actions
Introduce separate actions for creating login or shell sessions for
the local host or a local container. By default allow local unprivileged
clients to create new login sessions (which is safe, since getty will
ask for username and authentication).
Also, imply login privs from shell privs, as well as shell and login
privs from manage privs.
Lennart Poettering [Mon, 24 Aug 2015 19:09:49 +0000 (21:09 +0200)]
systemctl: properly handle empty control group paths in "status"
When showing the status of the "-.slice" slice root unit (whose reported
cgroup path is ""), we suppressed the cgroup tree so far, because
skipped it for all unit with an empty cgroup path. Let's fix that, and
properly handle the empty cgroup path.
Lennart Poettering [Mon, 24 Aug 2015 19:09:16 +0000 (21:09 +0200)]
machined: userns is only supported for container-class machines
We do not support userns for VM machines or for the host itself.
Lennart Poettering [Mon, 24 Aug 2015 19:08:04 +0000 (21:08 +0200)]
machinectl: don't show ".host" pseudo-machine in list by default
Let's hide all machines whose name begins with "." by default, thus
hiding the ".host" pseudo-machine, unless --all is specified. This
takes inspiration from the ".host" image handling in "machinectl
list-images" which also hides all images whose name starts with ".".
Lennart Poettering [Mon, 24 Aug 2015 19:05:09 +0000 (21:05 +0200)]
machined: introduce pseudo-machine ".host" refererring to the host system
Some of the operations machined/machinectl implement are also very
useful when applied to the host system (such as machinectl login,
machinectl shell or machinectl status), hence introduce a pseudo-machine
by the name of ".host" in machined that refers to the host system, and
may be used top execute operations on the host system with.
This copies the pseudo-image ".host" machined already implements for
image related commands.
(This commit also adds a PK privilege for opening a PTY in a container,
which was previously not accessible for non-root.)
Lennart Poettering [Sun, 23 Aug 2015 12:33:50 +0000 (14:33 +0200)]
machined: validate machine names at more places
When enumerating machines from /run, and when accepting machine names
for operations, be more strict and always validate.
Note that these checks are strictly speaking unnecessary, since
enumeration happens only on the trusted /run...
Lennart Poettering [Sun, 23 Aug 2015 12:30:52 +0000 (14:30 +0200)]
util: make machine_name_is_valid() a macro and move it to hostname-util.h
As it turns out machine_name_is_valid() does the exact same thing as
hostname_is_valid() these days, as it just invoked that and checked the
name length was < 64. However, hostname_is_valid() checks the length
against HOST_NAME_MAX anyway (which is 64 on Linux), hence any
additional check is redundant.
We hence replace machine_name_is_valid() by a macro that simply maps it
to hostname_is_valid() but sets the allow_trailing_dot parameter to
false. We also move this this call to hostname-util.h, to the same place
as the hostname_is_valid() declaration.
Lennart Poettering [Sun, 23 Aug 2015 12:29:59 +0000 (14:29 +0200)]
util: make hostname_is_valid() easier to read
Add more comments, and rename some parameters and variables to be more
expressive.
Lennart Poettering [Sun, 23 Aug 2015 12:04:31 +0000 (14:04 +0200)]
machined: always look for leader PID first
When looking for the machine belonging to a PID, always look for the
leader first, only then fall back to a cgroup check. We keep direct
track of the leader PID, but only indirectly of the cgroup, hence prefer
the PID.
Lennart Poettering [Sun, 23 Aug 2015 11:24:10 +0000 (13:24 +0200)]
machinectl: add new "machinectl shell" command
This makes use of machined's new OpenShell() command and allows opening
a new interactive shell in any container.
Lennart Poettering [Sun, 23 Aug 2015 11:20:58 +0000 (13:20 +0200)]
machined: add new OpenShell() bus call
This new bus call opens an interactive shell in a container. It works
like the existing OpenLogin() call, but does not involve getty, and
instead opens an arbitrary command line.
This is similar to "systemd-run -t -M" but is controlled by a specific
PolicyKit privilege.
Lennart Poettering [Sun, 23 Aug 2015 11:19:21 +0000 (13:19 +0200)]
core: open up more executable properties via the bus
This is preparation for a later commit that makes use of these
properties for spawning an interactive shell in a container.
Lennart Poettering [Sun, 23 Aug 2015 11:14:04 +0000 (13:14 +0200)]
core: optionally create LOGIN_PROCESS or USER_PROCESS utmp entries
When generating utmp/wtmp entries, optionally add both LOGIN_PROCESS and
INIT_PROCESS entries or even all three of LOGIN_PROCESS, INIT_PROCESS
and USER_PROCESS entries, instead of just a single INIT_PROCESS entry.
With this change systemd may be used to not only invoke a getty directly
in a SysV-compliant way but alternatively also a login(1) implementation
or even forego getty and login entirely, and invoke arbitrary shells in
a way that they appear in who(1) or w(1).
This is preparation for a later commit that adds a "machinectl shell"
operation to invoke a shell in a container, in a way that is compatible
with who(1) and w(1).
Tom Gundersen [Mon, 24 Aug 2015 16:37:02 +0000 (18:37 +0200)]
Merge pull request #1012 from gentoo-root/master
sd-device: fix enumeration of devices without subsystem
David Herrmann [Mon, 24 Aug 2015 11:41:03 +0000 (13:41 +0200)]
sd-bus: don't list activators as proper peers
If a connection passed KDBUS_HELLO_ACTIVATOR, it cannot do I/O on the
bus. Hence, we should not treat it as proper peer. To actually query it,
you have to explicitly ask for activators.
This makes kdbus in-line with what dbus-daemon does.
David Herrmann [Mon, 24 Aug 2015 10:56:37 +0000 (12:56 +0200)]
Revert "sd-bus: include queried path in GetManagedObjects"
This reverts commit
92d16a53e385781a55d9231d9f8f89c1747ab0e4. As it turns
out, this is not how ObjectManager is supposed to work. It is just a
special behavior of BlueZ, but no-one else implements it this way.
Revert the patch as discussed on github, and as such revert to the
previous behavior (as described in the spec).
Daniel Mack [Mon, 24 Aug 2015 08:46:59 +0000 (10:46 +0200)]
Merge pull request #1014 from whot/hwdb-updates
hwdb: add more DPI entries
Peter Hutterer [Mon, 24 Aug 2015 00:47:30 +0000 (10:47 +1000)]
hwdb: add more DPI entries
Provided by Francois Marier
Maxim Mikityanskiy [Sat, 22 Aug 2015 08:33:32 +0000 (11:33 +0300)]
sd-device: fix enumeration of devices without subsystem
Prior to commit
c32eb440bab953a0169cd207dfef5cad16dfb340, libudev's
function udev_enumerate_scan_devices() had behaved differently. If
parent match was added with udev_enumerate_add_match_parent(),
udev_enumerate_scan_devices() did not return error if some child devices
had no subsystem symlink in sysfs. An example of such devices is USB
endpoints /sys/bus/usb/devices/*/ep_*. If there was a parent match
against USB device, old implementation of udev_enumerate_scan_devices()
did not treat ep_* device directories without subsystem symlink as error
and just ignored them, but new implementation returns -ENOENT (also
ignoring these devices) though correctly enumerates all other matching
devices.
To compare, you could look at
96df036fe3d25525a44f5efdb2fc8560e82e6cfd,
in src/libudev/libudev-enumerate.c, function parent_add_child():
if (!match_subsystem(enumerate, udev_device_get_subsystem(dev)))
goto nomatch;
udev_device_get_subsystem() was returning NULL, match_subsystem() was
returning false, and USB endpoint device was ignored.
New parent_add_child() from src/libsystemd/sd-device/device-enumerator.c
checks return value of sd_device_get_subsystem() and fails if subsystem
was not found. Absence of subsystem symlink should not be really treated
as error because all enumerations of children of USB devices will fail
with -ENOENT. This new behavior also breaks system-config-printer.
So restore old behavior and treat absence of subsystem symlink as no
match.
Tom Gundersen [Fri, 21 Aug 2015 23:57:58 +0000 (01:57 +0200)]
Merge pull request #1010 from poettering/resolved-question-key
only maintain one question RR key per transaction and other fixes
Lennart Poettering [Fri, 21 Aug 2015 21:07:49 +0000 (23:07 +0200)]
Merge pull request #1009 from phomes/master
remove unused variables
Lennart Poettering [Fri, 21 Aug 2015 20:59:38 +0000 (22:59 +0200)]
resolved: always split up questions into per-RR transactions
We do so for Unicast DNS and LLMNR anyway, let's also do this for mDNS,
and simplify things.
Lennart Poettering [Fri, 21 Aug 2015 20:55:01 +0000 (22:55 +0200)]
resolved: only maintain one question RR key per transaction
Let's simplify things and only maintain a single RR key per transaction
object, instead of a full DnsQuestion. Unicast DNS and LLMNR don't
support multiple questions per packet anway, and Multicast DNS suggests
coalescing questions beyond a single dns query, across the whole system.
Lennart Poettering [Fri, 21 Aug 2015 20:51:05 +0000 (22:51 +0200)]
resolved: add extra check for family when doing LLMNR TCP connections
It shouldn't happen that we try to resolve IPv4 addresses via LLMNR on
IPv6 and vice versa, but let's explicitly verify that we don't turn an
IPv4 LLMNR lookup into an IPv6 TCP connection.
Lennart Poettering [Fri, 21 Aug 2015 20:47:06 +0000 (22:47 +0200)]
resolved: add reference to negative caching RFC
Thomas Hindoe Paaboel Andersen [Fri, 21 Aug 2015 20:17:48 +0000 (22:17 +0200)]
remove unused variables
Tom Gundersen [Fri, 21 Aug 2015 14:23:02 +0000 (16:23 +0200)]
Merge pull request #1005 from poettering/resolved-refuse-compression
Don't do name compression when passing RRs across the bus
Tom Gundersen [Fri, 21 Aug 2015 14:21:21 +0000 (16:21 +0200)]
Merge pull request #1004 from poettering/systemd-run-man
man: rework systemd-run man page a bit
Lennart Poettering [Fri, 21 Aug 2015 14:06:25 +0000 (16:06 +0200)]
resolve-host: Minor wording improvement
Lennart Poettering [Fri, 21 Aug 2015 14:04:59 +0000 (16:04 +0200)]
resolved: when passing RRs across the bus, make sure not to use name compression
We explicitly need to turn off name compression when marshalling or
demarshalling RRs for bus transfer, since they otherwise refer to packet
offsets that reference packets that are not transmitted themselves.
Lennart Poettering [Fri, 21 Aug 2015 14:04:16 +0000 (16:04 +0200)]
man: rework systemd-run man page a bit